Sync model repo (text/metadata)

.python-version

@@ -0,0 +1 @@
+3.12

README.md

@@ -1,218 +1,190 @@
 ---
-library_name: executorch
-license: mit
+license: apache-2.0
 pipeline_tag: automatic-speech-recognition
-base_model: openai/whisper-small
-base_model_relation: quantized
+library_name: executorch
 tags:
-  - automatic-speech-recognition
-  - whisper
-  - int8
-  - quantized
-  - xnnpack
   - arm
+  - arm-optimized
+  - premium-smartphone
   - executorch
-  - edge-ai
-  - librispeech
-datasets:
-  - librispeech_asr
-metrics:
-  - wer
-  - cer
-model-index:
-  - name: whisper-small-int8-xnnpack-executorch
-    results:
-      - task:
-          type: automatic-speech-recognition
-        dataset:
-          type: librispeech_asr
-          name: LibriSpeech ASR (test-clean)
-          split: test
-          args:
-            evaluation_samples: 2620
-        metrics:
-          - type: wer
-            value: 3.41
-            name: WER (ExecuTorch)
-          - type: cer
-            value: 1.29
-            name: CER (ExecuTorch)
+base_model: openai/whisper-small
+base_model_relation: quantized
 ---
 
-# Whisper Small INT8 (ExecuTorch + XNNPACK + KleidiAI)
+# Whisper Small optimized for Arm-based Premium Smartphone
 
-INT8 8da8w quantized version of [openai/whisper-small](https://huggingface.co/openai/whisper-small),
-optimized for ARM deployment using ExecuTorch, XNNPACK, and KleidiAI. The model targets
-on-device English speech-to-text transcription on Android and ARM-based edge hardware,
-delivering near-identical accuracy at substantially lower latency and memory footprint.
+Whisper Small, an encoder-decoder Transformer for automatic speech recognition, quantized to INT8 and exported to ExecuTorch for on-device inference on Arm-based Premium Smartphone devices.
 
-## Key Highlights
+## Summary
 
-Compared to the FP32 baseline on ARM hardware (Vivo X300, Arm C1):
+This repository contains an Arm-optimized version of openai/whisper-small for automatic speech recognition. The model is provided in ExecuTorch (`.pte`) format, targeting Premium Smartphone systems.
 
-- **2.72x smaller** — .pte artifact shrinks from 1074.76 MB to 395.05 MB
-- **1.40x faster inference** — end-to-end latency drops from 10927.0 ms to 7802.5 ms (p50)
-- **WER preserved** — 3.45% to 3.41% on LibriSpeech test-clean (2620 utterances)
+This version is intended to demonstrate efficient inference on Arm-based platforms while preserving the original model's intended behavior. Arm has evaluated this model on LibriSpeech ASR and measured performance on Vivo X300.
 
-## Model Details
+**Key results**
 
-| Property | Value |
+| Area | Result |
 |---|---|
-| Developed by | OpenAI |
-| Model type | Automatic Speech Recognition (encoder-decoder Transformer) |
-| Language | English |
-| License | MIT |
-| Base model | [openai/whisper-small](https://huggingface.co/openai/whisper-small) |
-| Modification | Post-training quantization (8da8w), not finetuned |
-| Parameter count | 241.73M |
-| PyTorch state dict size | 922.31 MB |
-| Optimized .pte size | 395.05 MB |
-
-## How to Get Started
+| Model format | ExecuTorch (`.pte`) |
+| Target device class | Premium Smartphone |
+| Reference device | Vivo X300 (C1-Ultra, C1-Premium, C1-Pro; Android 16 / OriginOS 6) |
+| Primary performance result | 7802.5 ms p50 latency, RTFx 0.90 |
+| Accuracy result | Normalised WER 3.41%, CER 1.29% |
+| Size / memory result | 395.05 MB, 2.72 x smaller than the baseline (1074.76 MB) |
 
-### Install dependencies
+## Original model
 
-```bash
-pip install transformers torch soundfile numpy
-```
+| Field | Value |
+|---|---|
+| Original model | openai/whisper-small |
+| Original source | Hugging Face |
+| Original developer | OpenAI |
+| Original model card | [openai/whisper-small](https://huggingface.co/openai/whisper-small) |
+| Original license | [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0) |
 
-### Run inference
+## Model files
 
-```bash
-python example.py
-```
+| File | Description |
+|---|---|
+| `whisper_small_vivo_executorch_optimized.pte` | Arm-optimized model for deployment |
+| `whisper_preprocessor.pte` | ExecuTorch module that computes the log-mel spectrogram from raw audio; loaded by `example.py` when present, with a Python-side fallback otherwise |
+| `example.py` | Minimal inference example |
+| `pyproject.toml` | Pinned runtime dependencies for `example.py`, resolved with uv |
+| `uv.lock` | Locked dependency resolution for `pyproject.toml` |
+| `config.yaml` | Model I/O contract used by the example |
+| `benchmarks/` | FP32 baseline and Arm-optimized benchmark records |
 
-### Core inference loop
-
-```python
-from transformers import WhisperForConditionalGeneration, WhisperProcessor
-import torch
-
-MODEL_NAME = "openai/whisper-small"
-LANGUAGE = "en"
-TASK = "transcribe"
-MAX_NEW_TOKENS = 128
-
-processor = WhisperProcessor.from_pretrained(MODEL_NAME)
-model = WhisperForConditionalGeneration.from_pretrained(MODEL_NAME)
-model.eval()
-
-def transcribe(audio_path: str) -> str:
-    """Load audio and return transcribed text."""
-    import soundfile as sf
-    import numpy as np
-
-    audio_np, sr = sf.read(audio_path, dtype="float32")
-    if audio_np.ndim == 2:
-        audio_np = audio_np.mean(axis=1)
-
-    features = processor.feature_extractor(
-        audio_np, sampling_rate=sr, return_tensors="pt"
-    ).input_features  # [1, 80, 3000]
-
-    with torch.no_grad():
-        output_ids = model.generate(
-            features,
-            language=LANGUAGE,
-            task=TASK,
-            max_new_tokens=MAX_NEW_TOKENS,
-        )
-    return processor.tokenizer.batch_decode(output_ids, skip_special_tokens=True)[0].strip()
-```
+## Performance
 
-> **Note on .pte artifacts:** The `pte_optimized/` and `pte_original/` directories contain
-> ExecuTorch-serialized models for on-device ARM inference. Running them requires the
-> ExecuTorch C++ seq2seq runner — they are not suitable for a simple Python
-> `method.execute()` call. The `example.py` script above uses the HuggingFace Transformers
-> library for Python-based evaluation and prototyping.
+Performance was measured on the reference configuration below. Results are intended to make the optimization reproducible but do not guarantee identical performance on every Arm-based system.
 
-## Evaluation
+**Reference configuration**
 
-### Testing data
+| Field | Value |
+|---|---|
+| Device / platform | Vivo X300 |
+| CPU / accelerator | C1-Ultra, C1-Premium, C1-Pro (aarch64, 8 cores), CPU execution backend |
+| OS | Android 16 / OriginOS 6 |
+| Runtime | ExecuTorch 1.1.0 |
+| Backend / delegate | XNNPACK, KleidiAI |
+| Batch size | 1 |
+| Precision | INT8 weights (per-channel symmetric) and INT8 dynamic activations (PTQ-dynamic, no calibration required) |
+| Runs | 10 warmup, 50 measured |
+
+**Performance results**
+
+| Metric | Original / baseline | Arm-optimized | Improvement |
+|---|---:|---:|---:|
+| p50 latency | 10927.0 ms | 7802.5 ms | 1.40 x faster |
+| p90 latency | 11742.0 ms | 8027.0 ms | 1.46 x faster |
+| p99 latency | 11742.0 ms | 8027.0 ms | 1.46 x faster |
+| Model size | 1074.76 MB | 395.05 MB | 2.72 x smaller |
+| Peak memory | 5226.97 MB | 4662.29 MB | 1.12 x less |
+| RTFx | 0.64 | 0.90 | 1.40 x higher |
+| Model load time | 1327.0 ms | 1015.0 ms | 1.31 x faster |
+
+## Accuracy
+
+Accuracy was evaluated using the same preprocessing, input resolution, and evaluation protocol described below. Where possible, the optimized model is compared against the original model under the same evaluation conditions.
+
+**Evaluation setup**
+
+| Field | Value |
+|---|---|
+| Dataset | LibriSpeech ASR (librispeech_asr) |
+| Split | test-clean |
+| Number of samples | 2620 |
+| Metric(s) | Normalised WER (Whisper-style), CER |
+| Evaluation runtime | ExecuTorch |
 
-2620 utterances from the LibriSpeech ASR `test-clean` split. Calibration: none required
-(dynamic activation quantization). Dynamic weight-only 8da8w — weights INT8 offline,
-activations INT8 quantized per-token at runtime.
+**Accuracy results**
 
-### Metrics
+| Metric | Original / baseline | Arm-optimized | Change |
+|---|---:|---:|---:|
+| Normalised WER | 3.45% | 3.41% | -0.04 pp |
+| CER | 1.33% | 1.29% | -0.04 pp |
 
-| Metric | Description |
-|---|---|
-| WER | Word Error Rate (lower is better) |
-| CER | Character Error Rate (lower is better) |
+Accuracy was measured using the evaluation setup described above. Users should re-evaluate the model on their own data before production use.
 
-### Accuracy results
+## Arm optimization approach
 
-| Model | WER | CER |
+Arm optimized this model for efficient inference on Arm-based platforms using a hardware-aware conversion and validation flow.
+
+For this release, Arm used:
+
+| Optimization area | Applied? | Notes |
 |---|---|---|
-| openai/whisper-small (FP32) | 3.45% | 1.33% |
-| whisper-small INT8 8da8w (ExecuTorch) | **3.41%** | **1.29%** |
+| Model conversion | Yes | Converted to ExecuTorch `.pte` via an Optimum ExecuTorch seq2seq export |
+| Quantization | Yes | PTQ-dynamic INT8: 8-bit weights (per-channel symmetric), 8-bit dynamic activations; no calibration required |
+| Runtime/backend selection | Yes | XNNPACK and KleidiAI optimisations |
+| Graph/runtime compatibility updates | Yes | Performed as part of the ExecuTorch export pipeline |
+| Accuracy validation | Yes | Compared against the original model or published baseline |
+| Performance validation | Yes | Measured on the reference Arm platform |
 
-### Efficiency results (Vivo X300, Arm C1)
+The goal of this process is to improve deployment characteristics such as latency, memory use, model size, and runtime compatibility while preserving the model's intended behavior. Detailed conversion scripts, calibration configuration, or backend-specific implementation details may be provided separately where appropriate.
 
-| Metric | FP32 Original | INT8 Optimized | Improvement |
-|---|---|---|---|
-| .pte file size | 1074.76 MB | 395.05 MB | **2.72x smaller** |
-| End-to-end latency p50 | 10927.0 ms | 7802.5 ms | **1.40x faster** |
-| End-to-end latency p90 | 11742.0 ms | 8027.0 ms | 1.46x faster |
-| RTFx | 0.641 | 0.897 | **+40%** |
-| Decode throughput | 3.62 tok/s | 5.69 tok/s | **+57%** |
-| Prefill throughput | 620.0 tok/s | 745.6 tok/s | +20% |
-| Time to first token (TTFT) | 2683.9 ms | 2196.2 ms | -18% |
-| Peak memory (USS) | 5226.97 MB | 4662.29 MB | **1.12x less** |
-| Model load time | 1327.3 ms | 1015.0 ms | 1.31x faster |
+## Using this model
 
-Benchmark conditions: batch size 1, ~7 s audio clips, 50 runs (10 warmup), offline mode,
-16 kHz input on Android 16 / OriginOS 6.
+### Install dependencies
 
-## Technical Specifications
+Dependencies are declared in `pyproject.toml`, which ships with this repository. Resolve and install them into a local virtual environment with uv:
 
-### Objective
+```bash
+uv python install
+uv sync --frozen
+```
 
-English speech-to-text transcription using Whisper's encoder-decoder Transformer
-architecture. The encoder processes 80-bin log-mel spectrograms; the decoder
-autoregressively generates token IDs which are decoded to text with the Whisper tokenizer.
+### Run the example
 
-### Quantization
+```bash
+uv run example.py
+```
 
-- **Method:** TorchAO 8da8w — 8-bit dynamic activation quantization + 8-bit weight quantization
-- **Calibration:** none required (dynamic activation quantization). Dynamic weight-only 8da8w — weights INT8 offline, activations INT8 quantized per-token at runtime.
-- **Weight granularity:** per-channel
-- **Symmetry:** symmetric INT8
-- **Skipped layers (kept FP32):** `proj_out`/`lm_head`, encoder + decoder positional embeddings, `encoder.conv1` / `encoder.conv2`
+Note: The Python/uv example runs on AWS Graviton (Ubuntu arm64) to confirm runtime compatibility only, and is intended as a guideline for building an equivalent run script on smartphone devices.
 
-### Export pipeline
+### Expected input
 
-1. Load pretrained FP32 `openai/whisper-small` from HuggingFace
-2. Export encoder and decoder to ExecuTorch Seq2Seq format via Optimum ExecuTorch
-3. Apply 8da8w (8-bit dynamic activation, 8-bit weight) quantization with TorchAO
-4. Export tokenizer files (`tokenizer.json`, `tokenizer_config.json`, `special_tokens_map.json`)
-5. Export audio preprocessor to `whisper_preprocessor.pte`
+| Property | Value |
+|---|---|
+| Input shape | [1, 80, 3000] |
+| Input type | float32 |
+| Input range | N/A (log-mel spectrogram magnitude, not a fixed bounded range) |
+| Preprocessing | Load audio as a 16 kHz mono waveform; compute a log-mel spectrogram (mel bins 80, hop length 160, n_fft 400, sample rate 16000, duration 30 seconds); pad or trim to 3000 time frames |
 
-### Input preprocessing
+### Expected output
 
-| Step | Parameters |
+| Property | Value |
 |---|---|
-| Load audio as waveform | 16 kHz mono |
-| Log-mel spectrogram | n_mels=80, hop_length=160, n_fft=400, sample_rate=16000, duration=30s |
-| Pad or trim | 3000 time frames |
+| Output shape | N/A (variable-length token ID sequence, autoregressive generation) |
+| Output type | Token ID sequence |
+| Postprocessing | Greedy decoding with a fixed token-suppression list and a repetition-guard heuristic that trims repeated trailing token patterns; decode with the Whisper tokenizer, skipping special tokens |
+
+## Intended use
+
+This model is intended for developers evaluating automatic speech recognition workloads on Arm-based platforms. It is suitable as a reference implementation for benchmarking, prototyping, and integration exploration.
+
+## Limitations
+
+- Performance depends on the target device, runtime version, backend/delegate support, memory configuration, and system load.
+- Accuracy was evaluated on LibriSpeech ASR and may not generalize to all domains.
+- This release preserves the original model's intended task and behavior, but users should validate it for their own application, data, and deployment environment.
+- This repository is not a replacement for the original model documentation.
+
+## Additional notes
+
+Quantization keeps a small set of layers in FP32 to preserve accuracy: proj_out/lm_head, the encoder and decoder positional embeddings, and encoder.conv1/encoder.conv2. Audio inputs are limited to 30 seconds (3000 mel time frames) per chunk; longer audio must be chunked externally before inference.
+
+`example.py` forces English transcription by hardcoding the decoder prefix to `<|en|>, <|transcribe|>, <|notimestamps|>`. This is an example-level default, not a model restriction: the bundled tokenizer and decoder support the full multilingual Whisper vocabulary (98 language tokens) and the `<|translate|>` task, so other languages or the translate task can be enabled by changing the forced-prefix token IDs in `example.py`, with no re-export required.
+
+## About this version
 
-**Input tensor:** `[1, 80, 3000]`, dtype `float32` — log-mel spectrogram (batch=1, mel bins, time frames)
+This repository contains a converted version of the openai/whisper-small model, originally developed by OpenAI.
+Arm has converted the model to enable efficient execution on Arm-based platforms. No changes have been made to the model's intended behavior.
 
-### Output postprocessing
+## Original model and documentation
 
-1. Autoregressive greedy decoding with the Whisper tokenizer
-2. Skip special tokens
+For full details of the original model, please refer to the original [model card](https://huggingface.co/openai/whisper-small).
 
-## Known Limitations
+## Purpose of this release
 
-- Evaluated on English speech (LibriSpeech test-clean); performance on other languages or
-  accents has not been measured for this INT8 variant.
-- The `.pte` runtime on Python requires the ExecuTorch C++ seq2seq runner — not a simple
-  `method.execute()` call. Use `example.py` (HuggingFace Transformers) for Python prototyping.
-- Maximum audio duration: 30 seconds per chunk (3000 mel time frames at 16 kHz).
-  Longer audio must be chunked externally.
-- Android latency measured on Vivo X300 with Arm C1 processor (1x C1-Ultra, 3x C1-Premium,
-  4x C1-Pro cores at 4.21 / 3.5 / 2.7 GHz). Results may differ on other ARM devices.
-- RTFx > 1.0 indicates real-time capable transcription on this hardware; values below 1.0
-  on lower-tier devices are expected.
+This version is provided by Arm as a reference implementation to demonstrate performance on Arm-based systems. It is not a production-ready or supported solution. Users should evaluate the model independently for their use-case. Arm provides no warranties or ongoing support for this version.

example.py

@@ -1,128 +1,385 @@
-"""Whisper Small ASR — inference example using HuggingFace Transformers.
+"""Minimal inference example for Whisper Small INT8 using ExecuTorch.
 
-The optimized.pte and original.pte artifacts in the sibling pte_optimized/ and
-pte_original/ directories are ExecuTorch-optimized models (8da8w INT8 dynamic
-quantization) intended for on-device ARM inference (Android / Graviton).
-Running those artifacts directly requires the ExecuTorch C++ runtime and a
-specialized seq2seq runner — they are not suitable for a simple Python
-``method.execute()`` call.
+Loads a quantized .pte model and transcribes a single audio file.
+The INT8 model was exported via Optimum-ExecuTorch with 8da8w quantization on
+all Linear layers plus a manual weight-only INT8 pass on `decoder.embed_tokens`,
+and uses separate 'encoder' and 'text_decoder' ExecuTorch methods with a
+static KV cache.
 
-This script demonstrates equivalent inference using the HuggingFace Transformers
-library, which is the recommended path for Python-based evaluation and prototyping.
+This example resolves the tokenizer and preprocessor artifacts from the same
+directory as the model, matching the benchmark flow used by the Whisper runner.
 
 Requirements:
-    pip install transformers torch soundfile numpy
+    pip install executorch torch transformers soundfile numpy
 """
 
+import argparse
 import json
+import time
 from pathlib import Path
 
+import numpy as np
 import torch
-import torch.nn.functional as F
-from transformers import WhisperForConditionalGeneration, WhisperProcessor
+from executorch.runtime import Runtime
+from transformers import AutoTokenizer
 
-# ── Configuration ──────────────────────────────────────────────────────────────
+# -- Configuration -------------------------------------------------------------
 AUDIO_PATH = "sample_input.flac"
-MODEL_NAME = "openai/whisper-small"
-LANGUAGE = "en"
-TASK = "transcribe"
-MAX_NEW_TOKENS = 128
-SAMPLE_RATE = 16000
-
-
-# ── Audio Loading ──────────────────────────────────────────────────────────────
-def load_audio(audio_path: str) -> tuple[torch.Tensor, int]:
-    """Load audio file and return (waveform_1d_float32, sample_rate)."""
-    try:
-        import soundfile as sf
-
-        audio_np, sr = sf.read(audio_path, dtype="float32")
-        if audio_np.ndim == 2:
-            audio_np = audio_np.mean(axis=1)
-        import numpy as np
-
-        return torch.from_numpy(audio_np.astype(np.float32)), int(sr)
-    except ImportError:
-        import torchaudio
-
-        waveform, sr = torchaudio.load(audio_path)
-        if waveform.shape[0] > 1:
-            waveform = waveform.mean(dim=0)
-        else:
-            waveform = waveform.squeeze(0)
-        return waveform.float(), int(sr)
+PREPROCESSOR_FILENAME = "whisper_preprocessor.pte"
+DEFAULT_LOCAL_MODEL_DIR = "pte_optimized"
+MODEL_FILENAME = "whisper_small_vivo_executorch_optimized.pte"
+
+DECODER_START_TOKEN_ID = 50258
+FORCED_PREFIX_IDS = [50259, 50359, 50363]  # <|en|>, <|transcribe|>, <|notimestamps|>
+EOS_TOKEN_ID = 50257
+
+MAX_GENERATION_TOKENS = 128
+MAX_SECONDS_PER_SAMPLE = 120.0
+REPETITION_GUARD_REPEATS = 3
+REPETITION_GUARD_MIN_PATTERN_LEN = 2
+REPETITION_GUARD_MAX_PATTERN_LEN = 16
+
+SUPPRESS_TOKENS = (
+    1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63,
+    90, 91, 92, 93, 357, 366, 438, 532, 685, 705, 796, 930, 1058, 1220, 1267,
+    1279, 1303, 1343, 1377, 1391, 1635, 1782, 1875, 2162, 2361, 2488, 3467,
+    4008, 4211, 4600, 4808, 5299, 5855, 6329, 7203, 9609, 9959, 10563, 10786,
+    11420, 11709, 11907, 13163, 13697, 13700, 14808, 15306, 16410, 16791,
+    17992, 19203, 19510, 20724, 22305, 22935, 27007, 30109, 30420, 33409,
+    34949, 40283, 40493, 40549, 47282, 49146, 50359, 50360, 50361,
+)
+BEGIN_SUPPRESS_TOKENS = (220, 50257)
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(
+        description="Run Whisper Small ExecuTorch inference from an exported model bundle."
+    )
+    parser.add_argument(
+        "--model-dir",
+        default=None,
+        help=(
+            "Directory containing the ExecuTorch model, tokenizer files, and optionally "
+            "whisper_preprocessor.pte. Defaults to the local huggingface bundle."
+        ),
+    )
+    parser.add_argument(
+        "--audio",
+        default=AUDIO_PATH,
+        help="Path to the input audio file (.flac/.wav).",
+    )
+    return parser.parse_args()
+
 
+def resolve_model_dir(script_dir: Path, requested_dir: str | None) -> Path:
+    candidates: list[Path] = []
+    if requested_dir:
+        candidates.append(Path(requested_dir))
+    candidates.append(script_dir / DEFAULT_LOCAL_MODEL_DIR)
 
-# ── Resampling ──────────────────────────────────────��──────────────────────────
-def resample_to_16k(waveform: torch.Tensor, sample_rate: int) -> torch.Tensor:
-    """Resample waveform to 16 kHz using linear interpolation."""
-    if sample_rate == SAMPLE_RATE:
+    for candidate in candidates:
+        bundle_dir = candidate.resolve()
+        if (bundle_dir / MODEL_FILENAME).exists():
+            return bundle_dir
+
+    searched = "\n".join(f"- {candidate.resolve()}" for candidate in candidates)
+    raise FileNotFoundError(
+        "Could not find a Whisper Small ExecuTorch model bundle. Searched:\n"
+        f"{searched}"
+    )
+
+
+def load_audio(audio_path: str) -> tuple[np.ndarray, int]:
+    import soundfile as sf
+
+    waveform, sample_rate = sf.read(str(audio_path), dtype="float32")
+    if waveform.ndim == 2:
+        waveform = waveform.mean(axis=1)
+    return waveform, int(sample_rate)
+
+
+def resample_to_16k(waveform: np.ndarray, sample_rate: int) -> np.ndarray:
+    if sample_rate == 16000:
         return waveform
-    new_len = int(round(len(waveform) * SAMPLE_RATE / sample_rate))
-    return F.interpolate(
-        waveform.view(1, 1, -1), size=new_len, mode="linear", align_corners=False
-    ).view(-1)
-
-
-# ── Preprocessing ──────────────────────────────────────────────────────────────
-def preprocess(audio_path: str, processor: WhisperProcessor) -> torch.Tensor:
-    """Load audio and extract 80-bin log-mel spectrogram features [1, 80, 3000]."""
-    waveform, sr = load_audio(audio_path)
-    waveform_16k = resample_to_16k(waveform, sr)
-    features = processor.feature_extractor(
-        waveform_16k.numpy(), sampling_rate=SAMPLE_RATE, return_tensors="pt"
-    ).input_features
-    return features  # [1, 80, 3000]
-
-
-# ── Inference ──────────────────────────────────────────────────────────────────
-def transcribe(
-    audio_path: str,
-    model: WhisperForConditionalGeneration,
-    processor: WhisperProcessor,
-) -> str:
-    """Run Whisper inference and return transcribed text."""
-    features = preprocess(audio_path, processor)
+    target_len = int(round(len(waveform) * 16000 / sample_rate))
+    resampled = np.interp(
+        np.linspace(0, len(waveform) - 1, target_len),
+        np.arange(len(waveform)),
+        waveform,
+    )
+    return resampled.astype(np.float32)
+
+
+def load_preprocessor(preprocessor_path: Path):
+    if not preprocessor_path.exists():
+        return None, None
+
+    runtime = Runtime.get()
+    program = runtime.load_program(str(preprocessor_path))
+    method_names = sorted(program.method_names)
+    method_name = "forward" if "forward" in method_names else method_names[0]
+    return program, program.load_method(method_name)
+
+
+def preprocess(audio_path: str, preprocessor_method) -> torch.Tensor:
+    waveform, sample_rate = load_audio(audio_path)
+    waveform_16k = resample_to_16k(waveform, sample_rate)
+    waveform_tensor = torch.from_numpy(waveform_16k).float().contiguous()
+
+    if preprocessor_method is not None:
+        outputs = preprocessor_method.execute([waveform_tensor])
+        features = outputs[0]
+        if isinstance(features, (list, tuple)):
+            features = features[0]
+        return torch.as_tensor(features).float().contiguous()
+
+    from executorch.extension.audio.mel_spectrogram import WhisperAudioProcessor
+
+    fallback_preprocessor = WhisperAudioProcessor(
+        feature_size=80,
+        max_audio_len=300,
+        stack_output=True,
+    )
     with torch.no_grad():
-        output_ids = model.generate(
-            features,
-            language=LANGUAGE,
-            task=TASK,
-            max_new_tokens=MAX_NEW_TOKENS,
-        )
-    return processor.tokenizer.batch_decode(output_ids, skip_special_tokens=True)[0].strip()
+        features = fallback_preprocessor(waveform_tensor)
+    return features.float().contiguous()
+
+
+def load_model(pte_path: str) -> tuple:
+    runtime = Runtime.get()
+    program = runtime.load_program(pte_path)
+    available = sorted(program.method_names)
+    print(f"  Available methods: {available}")
+
+    if "encoder" in available and "text_decoder" in available:
+        return program, {
+            "format": "seq2seq",
+            "encoder": program.load_method("encoder"),
+            "decoder": program.load_method("text_decoder"),
+        }
+    if "forward" in available:
+        return program, {
+            "format": "forward",
+            "forward": program.load_method("forward"),
+        }
+    raise RuntimeError(f"Unknown export format. Methods found: {available}")
+
+
+def apply_suppression(scores: torch.Tensor, first_free_step: bool) -> torch.Tensor:
+    vocab_size = scores.shape[-1]
+    out = scores.clone()
+    valid_suppress = [
+        t for t in SUPPRESS_TOKENS
+        if 0 <= t < vocab_size and t != EOS_TOKEN_ID
+    ]
+    if valid_suppress:
+        out[0, valid_suppress] = float("-inf")
+    if first_free_step:
+        valid_begin = [t for t in BEGIN_SUPPRESS_TOKENS if 0 <= t < vocab_size]
+        if valid_begin:
+            out[0, valid_begin] = float("-inf")
+    return out
+
 
+def decode_tokens(tokenizer, token_ids: list[int]) -> str:
+    return tokenizer.decode(
+        token_ids,
+        skip_special_tokens=True,
+        clean_up_tokenization_spaces=False,
+    ).strip()
 
-# ── Save Results ───────────────────────────────────────────────────────────────
-def save_results(audio_path: str, text: str) -> None:
-    """Save transcription result to JSON in the same directory as this script."""
-    out_dir = Path(__file__).parent
-    result = {
-        "audio_file": str(Path(audio_path).name),
+
+def find_repeated_suffix_pattern(
+    token_ids: list[int],
+    *,
+    repeats: int = REPETITION_GUARD_REPEATS,
+    min_pattern_len: int = REPETITION_GUARD_MIN_PATTERN_LEN,
+    max_pattern_len: int = REPETITION_GUARD_MAX_PATTERN_LEN,
+) -> int | None:
+    total = len(token_ids)
+    upper = min(max_pattern_len, total // repeats)
+    for pattern_len in range(min_pattern_len, upper + 1):
+        pattern = token_ids[-pattern_len:]
+        if all(
+            token_ids[-pattern_len * (idx + 1) : -pattern_len * idx or None] == pattern
+            for idx in range(repeats)
+        ):
+            return pattern_len
+    return None
+
+
+def transcribe_seq2seq(
+    encoder_method,
+    decoder_method,
+    features: torch.Tensor,
+    tokenizer,
+) -> dict:
+    encoder_outputs = encoder_method.execute([features])
+    encoder_hidden = encoder_outputs[0]
+    if isinstance(encoder_hidden, (list, tuple)):
+        encoder_hidden = encoder_hidden[0]
+    encoder_hidden = torch.as_tensor(encoder_hidden).float().contiguous()
+
+    forced_prefix = list(FORCED_PREFIX_IDS)
+    tokens = [DECODER_START_TOKEN_ID]
+    cache_position = 0
+    forced_prefix_idx = 0
+    generated_token_count = 0
+    stop_reason = "max_tokens"
+    started = time.perf_counter()
+    generated_free_tokens: list[int] = []
+
+    for _step in range(MAX_GENERATION_TOKENS + len(forced_prefix)):
+        input_tensor = torch.tensor([[tokens[-1]]], dtype=torch.long).contiguous()
+        pos_tensor = torch.tensor([cache_position], dtype=torch.long).contiguous()
+
+        decoder_outputs = decoder_method.execute([input_tensor, encoder_hidden, pos_tensor])
+        flat_logits = decoder_outputs[0]
+        if isinstance(flat_logits, (list, tuple)):
+            flat_logits = flat_logits[0]
+        flat_logits = torch.as_tensor(flat_logits).float().flatten()
+
+        if forced_prefix_idx < len(forced_prefix):
+            next_token = forced_prefix[forced_prefix_idx]
+            forced_prefix_idx += 1
+        else:
+            scores = flat_logits.unsqueeze(0)
+            first_free = generated_token_count == 0
+            scores = apply_suppression(scores, first_free_step=first_free)
+            next_token = int(scores[0].argmax().item())
+            generated_token_count += 1
+            generated_free_tokens.append(next_token)
+
+            if next_token == EOS_TOKEN_ID:
+                stop_reason = "eos"
+                tokens.append(next_token)
+                cache_position += 1
+                break
+            repeated_suffix_len = find_repeated_suffix_pattern(generated_free_tokens)
+            if repeated_suffix_len is not None:
+                trim_count = repeated_suffix_len * REPETITION_GUARD_REPEATS
+                del generated_free_tokens[-trim_count:]
+                del tokens[-(trim_count - 1) :]
+                generated_token_count -= trim_count
+                stop_reason = "repetition_guard"
+                break
+            if time.perf_counter() - started >= MAX_SECONDS_PER_SAMPLE:
+                stop_reason = "timeout"
+                break
+
+        tokens.append(next_token)
+        cache_position += 1
+
+    elapsed = time.perf_counter() - started
+    text = decode_tokens(tokenizer, tokens)
+    return {
         "transcription": text,
-        "model": MODEL_NAME,
-        "language": LANGUAGE,
-        "task": TASK,
+        "generated_tokens": generated_token_count,
+        "stop_reason": stop_reason,
+        "elapsed_s": round(elapsed, 3),
     }
-    out_path = out_dir / "transcription.json"
-    with open(out_path, "w") as f:
-        json.dump(result, f, indent=2)
-    print(f"Transcription saved to: {out_path}")
 
 
-# ── Main ───────────────────────────────────────────────────────────────────────
+def transcribe_forward(forward_method, features: torch.Tensor, tokenizer) -> dict:
+    prompt = [DECODER_START_TOKEN_ID] + list(FORCED_PREFIX_IDS)
+    decoder_ids = torch.tensor([prompt], dtype=torch.long).contiguous()
+    generated_token_count = 0
+    stop_reason = "max_tokens"
+    started = time.perf_counter()
+    generated_free_tokens: list[int] = []
+
+    with torch.no_grad():
+        for _step in range(MAX_GENERATION_TOKENS):
+            outputs = forward_method.execute([features, decoder_ids])
+            logits = outputs[0]
+            if isinstance(logits, (list, tuple)):
+                logits = logits[0]
+            logits = torch.as_tensor(logits).float()
+            next_token_scores = logits[:, -1, :]
+            first_free = generated_token_count == 0
+            next_token_scores = apply_suppression(next_token_scores, first_free_step=first_free)
+            next_token = next_token_scores.argmax(dim=-1, keepdim=True).long()
+            decoder_ids = torch.cat([decoder_ids, next_token], dim=1)
+            generated_token_count += 1
+            generated_free_tokens.append(int(next_token.item()))
+
+            if EOS_TOKEN_ID >= 0 and bool(torch.all(next_token == EOS_TOKEN_ID)):
+                stop_reason = "eos"
+                break
+            repeated_suffix_len = find_repeated_suffix_pattern(generated_free_tokens)
+            if repeated_suffix_len is not None:
+                trim_count = repeated_suffix_len * REPETITION_GUARD_REPEATS
+                generated_free_tokens = generated_free_tokens[:-trim_count]
+                decoder_ids = decoder_ids[:, :-trim_count]
+                generated_token_count -= trim_count
+                stop_reason = "repetition_guard"
+                break
+            if time.perf_counter() - started >= MAX_SECONDS_PER_SAMPLE:
+                stop_reason = "timeout"
+                break
+
+    elapsed = time.perf_counter() - started
+    text = decode_tokens(tokenizer, decoder_ids[0].tolist())
+    return {
+        "transcription": text,
+        "generated_tokens": generated_token_count,
+        "stop_reason": stop_reason,
+        "elapsed_s": round(elapsed, 3),
+    }
+
+
+def save_results(result: dict, script_dir: Path) -> None:
+    output_path = script_dir / "transcription.json"
+    with open(output_path, "w", encoding="utf-8") as f:
+        json.dump(result, f, indent=2, ensure_ascii=False)
+    print(f"Saved transcription to {output_path}")
+
+
 def main() -> None:
-    print(f"Loading model: {MODEL_NAME}")
-    processor = WhisperProcessor.from_pretrained(MODEL_NAME)
-    model = WhisperForConditionalGeneration.from_pretrained(MODEL_NAME)
-    model.eval()
+    args = parse_args()
+    script_dir = Path(__file__).parent
+    model_dir = resolve_model_dir(script_dir, args.model_dir)
+    model_path = model_dir / MODEL_FILENAME
+    audio_arg = Path(args.audio)
+    audio_path = audio_arg if audio_arg.is_absolute() else (script_dir / audio_arg).resolve()
+    preprocessor_path = model_dir / PREPROCESSOR_FILENAME
+
+    print(f"Loading tokenizer from {model_dir} ...")
+    tokenizer = AutoTokenizer.from_pretrained(
+        str(model_dir),
+        local_files_only=True,
+        use_fast=True,
+    )
+
+    _preprocessor_program = None
+    preprocessor_method = None
+    if preprocessor_path.exists():
+        print(f"Loading preprocessor from {preprocessor_path} ...")
+        _preprocessor_program, preprocessor_method = load_preprocessor(preprocessor_path)
+    else:
+        print("Local preprocessor .pte not found; falling back to WhisperAudioProcessor.")
+
+    print(f"Loading model from {model_path} ...")
+    program, methods = load_model(str(model_path))
+
+    print(f"Preprocessing audio: {audio_path}")
+    features = preprocess(str(audio_path), preprocessor_method)
+    print(f"  Input features shape: {tuple(features.shape)}")
+
+    print("Running transcription ...")
+    if methods["format"] == "seq2seq":
+        result = transcribe_seq2seq(
+            methods["encoder"], methods["decoder"], features, tokenizer
+        )
+    else:
+        result = transcribe_forward(methods["forward"], features, tokenizer)
 
-    audio_path = str(Path(__file__).parent / AUDIO_PATH)
-    print(f"Transcribing: {audio_path}")
-    text = transcribe(audio_path, model, processor)
+    print(f"\nTranscription: {result['transcription']!r}")
+    print(f"Generated tokens: {result['generated_tokens']}")
+    print(f"Stop reason: {result['stop_reason']}")
+    print(f"Elapsed: {result['elapsed_s']:.3f} s")
 
-    print(f"\nTranscription: {text}")
-    save_results(audio_path, text)
+    save_results(result, script_dir)
 
 
 if __name__ == "__main__":

metadata.yaml

@@ -8,8 +8,10 @@ description: >-
   INT8 linear layers with selective FP32 components to reduce size and improve inference efficiency
   while preserving transcription quality.
 id: Arm/whisper-small-int8-xnnpack-executorch
-filename: whisper-small-int8-executorch.pte
+filename: whisper_small_vivo_executorch_optimized.pte
 base_model_id: openai/whisper-small
+vendor: OpenAI
+base_model_url: https://huggingface.co/openai/whisper-small
 profile: Arm-Optimized
 weight_dtype: int8
 quantization:

pyproject.toml

@@ -0,0 +1,38 @@
+[project]
+name = "whisper-small-int8-xnnpack-executorch-runtime"
+version = "0.1.0"
+description = "Runtime dependencies for example.py (Whisper Small INT8, ExecuTorch + XNNPACK)"
+requires-python = ">=3.12,<3.13"
+dependencies = [
+    "executorch==1.1.0",
+    "torch==2.10.0",
+    "transformers==4.57.1",
+    "soundfile==0.13.1",
+    "numpy==2.5.2",
+]
+
+[tool.uv]
+package = false
+# The example runs on Arm-based Linux; ExecuTorch and its Vivo X300 export target
+# aarch64. Restricting the resolution environment keeps the lock to the wheels
+# that platform actually installs.
+environments = ["sys_platform == 'linux' and platform_machine == 'aarch64'"]
+# coremltools targets Apple Core ML, which this XNNPACK-on-Arm-Linux example
+# never uses. Nothing in example.py imports it.
+exclude-dependencies = [
+    { package = { name = "executorch", version = "1.1.0" }, dependencies = ["coremltools"] },
+]
+
+[tool.ai-portal.deployment]
+schema-version = "1"
+runtime = "executorch"
+required-capabilities = ["XnnpackBackend"]
+
+[tool.ai-portal.deployment.ubuntu]
+packages = []
+
+[tool.ai-portal.deployment.raspbian]
+packages = []
+
+[tool.ai-portal.deployment.files]
+expected = ["transcription.json"]

transcription.json

@@ -1,6 +1,6 @@
 {
-  "transcription": "I thank all who have loved me in their hearts, with thanks and love from mine.",
-  "generated_tokens": 19,
+  "transcription": "Everything around appeared solitary, and would have been silent but for the continued plashing of the fountain, and the whole scene still maintained the monastic illusion which the fancy of Waverly had conjured up.",
+  "generated_tokens": 43,
   "stop_reason": "eos",
-  "elapsed_s": 1.114
+  "elapsed_s": 2.891
 }

uv.lock

@@ -0,0 +1,724 @@
+version = 1
+revision = 3
+requires-python = "==3.12.*"
+resolution-markers = [
+    "platform_machine == 'aarch64' and sys_platform == 'linux'",
+]
+supported-markers = [
+    "platform_machine == 'aarch64' and sys_platform == 'linux'",
+]
+
+[manifest]
+excludes = [{ package = { name = "executorch", version = "1.1.0" }, dependencies = ["coremltools"] }]
+
+[[package]]
+name = "antlr4-python3-runtime"
+version = "4.9.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/3e/38/7859ff46355f76f8d19459005ca000b6e7012f2f1ca597746cbcd1fbfe5e/antlr4-python3-runtime-4.9.3.tar.gz", hash = "sha256:f224469b4168294902bb1efa80a8bf7855f24c99aef99cbefc1bcd3cce77881b", size = 117034, upload-time = "2021-11-06T17:52:23.524Z" }
+
+[[package]]
+name = "certifi"
+version = "2026.7.22"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/a3/c2/24167ea9858356b47a87a50d39908bfdb72ceeefe0041586e704e5376b3a/certifi-2026.7.22.tar.gz", hash = "sha256:741e2c3b351ddf169a738da9f2c048608ff7f2c5cc02f1ebc6b118bb090d5d55", size = 138112, upload-time = "2026-07-22T03:35:12.644Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0b/a7/71ac2cff56fec219ed242bb11b8efb69fcc4bec75db06fb7bfe35de520e6/certifi-2026.7.22-py3-none-any.whl", hash = "sha256:62f22742b58a1a33014a2b6b706588a8d7e2a88ae7bd1a6ebe8c992928483775", size = 136983, upload-time = "2026-07-22T03:35:11.276Z" },
+]
+
+[[package]]
+name = "cffi"
+version = "2.1.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pycparser", marker = "implementation_name != 'PyPy'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9e/ef/008a1939e372c06329a3fce4279c02f328488f3526744906eeec3da7ad5f/cffi-2.1.1.tar.gz", hash = "sha256:dd31f52ea1086513bb9df30f8fcee9b8918323ae067a3d5b78bc826a000712be", size = 530807, upload-time = "2026-08-03T21:21:18.939Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/44/de/f98430906df1545ffde0d543dd124a7a439bc2cd32b36b9c53f805df7333/cffi-2.1.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:68e62fe11f30d5ca8289242866f0a5291402d8529ca2178ab8afc5c9694ae890", size = 222389, upload-time = "2026-08-03T21:19:48.331Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/d2/6cd24ae3be000a634109c247d1475d62e5616d0dc78c82770942ec384248/cffi-2.1.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:208f941bb9d18e768138677f0a6d2ce01f590df56043dda1df1535ac57c88517", size = 225232, upload-time = "2026-08-03T21:19:53.109Z" },
+]
+
+[[package]]
+name = "charset-normalizer"
+version = "3.5.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e5/3f/143b048436775b0f76ac3eec145c019e8173ccc2885c8f20319b996d5e83/charset_normalizer-3.5.1.tar.gz", hash = "sha256:6117b84ea48435e5356dc737f5121485c30920ba43375fa7b434fd753df0eac3", size = 171764, upload-time = "2026-08-15T08:20:44.807Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9a/4c/be49ada26b1f0232d57aa89bbebf997a5cc2332a5616b6eca26ff680044d/charset_normalizer-3.5.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4582c27e8c889d64811987b5967fbd3ae0c823fe1fd933b543d55ac20bb475fa", size = 238530, upload-time = "2026-08-15T08:17:11.563Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/0b/c5292a2462d69b7378ea89793bbb5b2b6fcf6f7dd6d1667f9619094ad553/charset_normalizer-3.5.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:96fef3e886d6a9874b14f27fc193fbdc69d5d8035783d86aa4e1cea594e695f9", size = 240110, upload-time = "2026-08-15T08:17:20.547Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/c7/9e48cee5c161fe24da823b61bf381921d77cb994a0a4de148e95018c1984/charset_normalizer-3.5.1-cp37-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:cee5dd7c6fb5dd52a0fe2a740f9bc6e3593f5f8b1788bde49de02086f30182b2", size = 240930, upload-time = "2026-08-15T08:19:54.163Z" },
+    { url = "https://files.pythonhosted.org/packages/75/9c/019fbb9f4834491a160951349b1a3714439376f66e5f7cf18b4f18f0c7aa/charset_normalizer-3.5.1-cp37-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:a2028475ba855475b8b4d3cfeb4994269c967aea8b9892dfba907f4263a863a3", size = 241821, upload-time = "2026-08-15T08:20:02.321Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/61/d01fc49b8dea277640b55a9e15960dbca9fdc8c9fde18e572d39c59f4019/charset_normalizer-3.5.1-py3-none-any.whl", hash = "sha256:6df0ec430f9a831772c23ca5a224cba36517a58a84bb32c32bb59a9fa67c47f6", size = 68658, upload-time = "2026-08-15T08:20:43.306Z" },
+]
+
+[[package]]
+name = "execnet"
+version = "2.1.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/bf/89/780e11f9588d9e7128a3f87788354c7946a9cbb1401ad38a48c4db9a4f07/execnet-2.1.2.tar.gz", hash = "sha256:63d83bfdd9a23e35b9c6a3261412324f964c2ec8dcd8d3c6916ee9373e0befcd", size = 166622, upload-time = "2025-11-12T09:56:37.75Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ab/84/02fc1827e8cdded4aa65baef11296a9bbe595c474f0d6d758af082d849fd/execnet-2.1.2-py3-none-any.whl", hash = "sha256:67fba928dd5a544b783f6056f449e5e3931a5c378b128bc18501f7ea79e296ec", size = 40708, upload-time = "2025-11-12T09:56:36.333Z" },
+]
+
+[[package]]
+name = "executorch"
+version = "1.1.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "expecttest" },
+    { name = "flatbuffers" },
+    { name = "hydra-core" },
+    { name = "hypothesis" },
+    { name = "kgb" },
+    { name = "mpmath" },
+    { name = "numpy" },
+    { name = "omegaconf" },
+    { name = "packaging" },
+    { name = "pandas" },
+    { name = "parameterized" },
+    { name = "pytest" },
+    { name = "pytest-json-report" },
+    { name = "pytest-rerunfailures" },
+    { name = "pytest-xdist" },
+    { name = "pytorch-tokenizers" },
+    { name = "pyyaml" },
+    { name = "ruamel-yaml" },
+    { name = "scikit-learn" },
+    { name = "sympy" },
+    { name = "tabulate" },
+    { name = "torch" },
+    { name = "torchao" },
+    { name = "typing-extensions" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c4/85/c03b1eba2ac6fde9586087c265f3c97dd2b3f847d088d3a92d601dbb02e7/executorch-1.1.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:a0d8214c3fb260c3d8a039f67fcd9a4dc5dc0c2c9f8b3c6de8f91e5666cfd829", size = 12185514, upload-time = "2026-01-28T15:58:47.89Z" },
+]
+
+[[package]]
+name = "expecttest"
+version = "0.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/22/0a/8868083a2b25c28811a88da6a4cb561f84feab0c9fa953c45820a3eb3a65/expecttest-0.3.0.tar.gz", hash = "sha256:6e8512fb86523ada1f94fd1b14e280f924e379064bb8a29ee399950e513eeccd", size = 7742, upload-time = "2024-12-11T15:32:15.72Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/27/fb/deeefea1ea549273817ca7bed3db2f39cc238a75a745a20e3651619f7335/expecttest-0.3.0-py3-none-any.whl", hash = "sha256:60f88103086e1754240b42175f622be83b6ffeac419434691ee5a5be819d0544", size = 8238, upload-time = "2024-12-11T15:32:13.523Z" },
+]
+
+[[package]]
+name = "filelock"
+version = "3.32.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/6d/30/03b03951873a1a0ffc7e8ca0e10c15597b59e8d0e39260704cd2ea087bc4/filelock-3.32.4.tar.gz", hash = "sha256:2bde2e4cf732e0153406d8a7bc80620ecf5e621fe0d25e41143c4e3b4733ff30", size = 222126, upload-time = "2026-08-23T17:37:55.363Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/01/a4/9b63d595d748e3aff8812b65eacc1a2c4bd90b7c2012e08e72373b4835eb/filelock-3.32.4-py3-none-any.whl", hash = "sha256:22e58ca3b1ae3b98993b762d7338367ae64fe50252bf78d59da3bfebcdf1cedd", size = 99864, upload-time = "2026-08-23T17:37:53.913Z" },
+]
+
+[[package]]
+name = "flatbuffers"
+version = "25.12.19"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e8/2d/d2a548598be01649e2d46231d151a6c56d10b964d94043a335ae56ea2d92/flatbuffers-25.12.19-py2.py3-none-any.whl", hash = "sha256:7634f50c427838bb021c2d66a3d1168e9d199b0607e6329399f04846d42e20b4", size = 26661, upload-time = "2025-12-19T23:16:13.622Z" },
+]
+
+[[package]]
+name = "fsspec"
+version = "2026.7.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/00/78/f34251dadb8f3921264a1d9b8946f5e542014ee2614b285261b4e40e6775/fsspec-2026.7.0.tar.gz", hash = "sha256:c803c40f4cf860b49dea58ee3e1c33cb9c790520e233537e1340049f89b82a88", size = 317040, upload-time = "2026-07-28T16:34:51.052Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/fd/3c/6a2bf344106328fd04963664a60b9bb6496fc25df8e962fcdc1367285fb9/fsspec-2026.7.0-py3-none-any.whl", hash = "sha256:b57ddbafedfaef7018c1ecab32aa200a9d7ca26b77965f64e48b70061249d279", size = 206583, upload-time = "2026-07-28T16:34:49.538Z" },
+]
+
+[[package]]
+name = "hf-xet"
+version = "1.6.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/1b/ab/522a2ab67f27971a9d48ca666d4fca85ef7d5282d142e31fd087e27b1bbe/hf_xet-1.6.0.tar.gz", hash = "sha256:2e58454a340b3556dfa4972d5451aff4fba8dd42a236600ba1a1d2b1514f0fef", size = 920527, upload-time = "2026-08-03T22:33:13.243Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9a/69/1f0cbc2fb22ae6082d094f743d1b8945a3f36f6089cb95f42b7ee348cda7/hf_xet-1.6.0-cp38-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:0e6e21fa3cdfcdcd76748564bf593870a5e013f47d97cf10aed63aa222cff5b7", size = 4262538, upload-time = "2026-08-03T22:33:05.287Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/3a/4f4f2301ade26e404462d3336fa11f7958d914cabbabdd6e03c3c5d5658c/hf_xet-1.6.0-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:4fc74352a17015bd0ee90038bc9efe38db894cde45f268b6712b04fce8cd0acb", size = 4460520, upload-time = "2026-08-03T22:33:06.81Z" },
+]
+
+[[package]]
+name = "huggingface-hub"
+version = "0.36.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "filelock" },
+    { name = "fsspec" },
+    { name = "hf-xet" },
+    { name = "packaging" },
+    { name = "pyyaml" },
+    { name = "requests" },
+    { name = "tqdm" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/7c/b7/8cb61d2eece5fb05a83271da168186721c450eb74e3c31f7ef3169fa475b/huggingface_hub-0.36.2.tar.gz", hash = "sha256:1934304d2fb224f8afa3b87007d58501acfda9215b334eed53072dd5e815ff7a", size = 649782, upload-time = "2026-02-06T09:24:13.098Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a8/af/48ac8483240de756d2438c380746e7130d1c6f75802ef22f3c6d49982787/huggingface_hub-0.36.2-py3-none-any.whl", hash = "sha256:48f0c8eac16145dfce371e9d2d7772854a4f591bcb56c9cf548accf531d54270", size = 566395, upload-time = "2026-02-06T09:24:11.133Z" },
+]
+
+[[package]]
+name = "hydra-core"
+version = "1.3.5"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "antlr4-python3-runtime" },
+    { name = "omegaconf" },
+    { name = "packaging" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/3e/e4/69a522676faf88994d93d8a5e69e0666c61cae7f73d1bbcc483222023e74/hydra_core-1.3.5.tar.gz", hash = "sha256:71c441eabbde086062045e4d3fce9e26015244f1a4ac721cf3e444c7edf10633", size = 3264337, upload-time = "2026-08-05T18:33:21.394Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9c/97/f9d463a6f3c7d0955753eca5cbbf35b596ac471dd13fe357211a53fd37be/hydra_core-1.3.5-py3-none-any.whl", hash = "sha256:a3ff35b4ea6794e4c83d993016f4bde4ac35797ebe7a08f30e83ed9341880331", size = 155768, upload-time = "2026-08-05T18:33:19.834Z" },
+]
+
+[[package]]
+name = "hypothesis"
+version = "6.165.10"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "sortedcontainers" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/5c/e2/0fad246d2b6330e1f78479bfc566b5c22be82aee8a865cde9a08f648487d/hypothesis-6.165.10.tar.gz", hash = "sha256:68b45e09834cd80523cb1eb274463073c7a9af4e4ef7cff34d9615f355572d32", size = 503703, upload-time = "2026-08-16T22:56:15.404Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/db/52/6f0a9b7aab24b0635e2238f3fbddea5b54b17879ac813df42a3cc3384c5c/hypothesis-6.165.10-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:76a7be86d986223b9f1bdb7e7cbcdb048649901fdb956c598ef73bdab1786cd5", size = 1108009, upload-time = "2026-08-16T22:54:53.082Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/f9/df24eb28412f82465e2b7707f0ff1ec274d580bce389d4d9156617dc7bba/hypothesis-6.165.10-cp310-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:2d0e0f8263d34dd8fa3b39eaa9a50bba56a8470b3dd9ebf6672d10840abe063e", size = 1283402, upload-time = "2026-08-16T22:54:18.054Z" },
+    { url = "https://files.pythonhosted.org/packages/04/66/4c71c5be7a49d84b8c3a9278c1807c4c81181ab5474beb27df9d4c40dc0e/hypothesis-6.165.10-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f9ff356e97e3ab09db07c8b675efa67340103874a0bae7465acb83dad7a35f7f", size = 1106830, upload-time = "2026-08-16T22:55:10.389Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/8b/794ced36864825492ac3712d5acab5a257b4601e6a9dc2ccdd3937198f87/hypothesis-6.165.10-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:e9acb2c4d9cb532c3fedea74159f7b923c8c036328c9239b4049e7aa073bdd81", size = 1280780, upload-time = "2026-08-16T22:54:34.983Z" },
+    { url = "https://files.pythonhosted.org/packages/10/39/ef26fa79c1738dfe9cdb1a3584fb6717d26429ca6c9d011cc4fdf08130c2/hypothesis-6.165.10-cp315-abi3.abi3t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7730d8197086f65d8969a991d6728a1d420a51b19fea06535c896cb43a1e05d0", size = 1104876, upload-time = "2026-08-16T22:54:58.937Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/60/31d504e364134d60af23e5f6365db0da3cf4a51b3ed3d4836e5a2cff12cf/hypothesis-6.165.10-cp315-abi3.abi3t-musllinux_1_2_aarch64.whl", hash = "sha256:e1bbeb7c506b07ee0422cf9b2f7212fefa4240957f03526d38d27bc6743a0a48", size = 1278684, upload-time = "2026-08-16T22:55:22.971Z" },
+]
+
+[[package]]
+name = "idna"
+version = "3.19"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/5f/f7/abb373e5757eaec4b922b92f97ec8d6d7e057cf06778247604fbc4e7c3f3/idna-3.19.tar.gz", hash = "sha256:5e0811a4383b21dc5838069f801c4fb62113b7447663d2530d2bd6e77b49bf15", size = 215237, upload-time = "2026-08-18T05:14:24.27Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/57/b0/0e52c878c53f245edd3a11020f20979b3f490f245af532c7cae3027754b5/idna-3.19-py3-none-any.whl", hash = "sha256:815e7be7a7806d54abb586dc943addc79e8b2ee16915059658cbeff4b1b43bf4", size = 68550, upload-time = "2026-08-18T05:14:22.343Z" },
+]
+
+[[package]]
+name = "iniconfig"
+version = "2.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/72/34/14ca021ce8e5dfedc35312d08ba8bf51fdd999c576889fc2c24cb97f4f10/iniconfig-2.3.0.tar.gz", hash = "sha256:c76315c77db068650d49c5b56314774a7804df16fee4402c1f19d6d15d8c4730", size = 20503, upload-time = "2025-10-18T21:55:43.219Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/cb/b1/3846dd7f199d53cb17f49cba7e651e9ce294d8497c8c150530ed11865bb8/iniconfig-2.3.0-py3-none-any.whl", hash = "sha256:f631c04d2c48c52b84d0d0549c99ff3859c98df65b3101406327ecc7d53fbf12", size = 7484, upload-time = "2025-10-18T21:55:41.639Z" },
+]
+
+[[package]]
+name = "jinja2"
+version = "3.1.6"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "markupsafe" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/df/bf/f7da0350254c0ed7c72f3e33cef02e048281fec7ecec5f032d4aac52226b/jinja2-3.1.6.tar.gz", hash = "sha256:0137fb05990d35f1275a587e9aee6d56da821fc83491a0fb838183be43f66d6d", size = 245115, upload-time = "2025-03-05T20:05:02.478Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/62/a1/3d680cbfd5f4b8f15abc1d571870c5fc3e594bb582bc3b64ea099db13e56/jinja2-3.1.6-py3-none-any.whl", hash = "sha256:85ece4451f492d0c13c5dd7c13a64681a86afae63a5f347908daf103ce6d2f67", size = 134899, upload-time = "2025-03-05T20:05:00.369Z" },
+]
+
+[[package]]
+name = "joblib"
+version = "1.5.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/41/f2/d34e8b3a08a9cc79a50b2208a93dce981fe615b64d5a4d4abee421d898df/joblib-1.5.3.tar.gz", hash = "sha256:8561a3269e6801106863fd0d6d84bb737be9e7631e33aaed3fb9ce5953688da3", size = 331603, upload-time = "2025-12-15T08:41:46.427Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/7b/91/984aca2ec129e2757d1e4e3c81c3fcda9d0f85b74670a094cc443d9ee949/joblib-1.5.3-py3-none-any.whl", hash = "sha256:5fc3c5039fc5ca8c0276333a188bbd59d6b7ab37fe6632daa76bc7f9ec18e713", size = 309071, upload-time = "2025-12-15T08:41:44.973Z" },
+]
+
+[[package]]
+name = "kgb"
+version = "7.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e8/00/9e56dee65ec791a92348fb54e8ced08c4c4db494b0f58cfb34737d087fb4/kgb-7.3.tar.gz", hash = "sha256:b8af7e79cb8b0df5a2ec596010b8e5d014845cfaa9203577b85b99d4df192927", size = 62922, upload-time = "2025-12-11T23:56:24.911Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/eb/d6/1c81a1292fc50ad93d0b145f1c241ecb7d541fba4dcec7166e2e1d99f9cd/kgb-7.3-py2.py3-none-any.whl", hash = "sha256:0b300cd6d234a951f60e54ccda78c99a355393d6ae878d3d5925e726ae2f0450", size = 59662, upload-time = "2025-12-11T23:56:23.699Z" },
+]
+
+[[package]]
+name = "markupsafe"
+version = "3.0.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/7e/99/7690b6d4034fffd95959cbe0c02de8deb3098cc577c67bb6a24fe5d7caa7/markupsafe-3.0.3.tar.gz", hash = "sha256:722695808f4b6457b320fdc131280796bdceb04ab50fe1795cd540799ebe1698", size = 80313, upload-time = "2025-09-27T18:37:40.426Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1e/2c/799f4742efc39633a1b54a92eec4082e4f815314869865d876824c257c1e/markupsafe-3.0.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3a7e8ae81ae39e62a41ec302f972ba6ae23a5c5396c8e60113e9066ef893da0d", size = 24332, upload-time = "2025-09-27T18:36:32.813Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/2f/336b8c7b6f4a4d95e91119dc8521402461b74a485558d8f238a68312f11c/markupsafe-3.0.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:be8813b57049a7dc738189df53d69395eba14fb99345e0a5994914a3864c8a4b", size = 23760, upload-time = "2025-09-27T18:36:36.001Z" },
+]
+
+[[package]]
+name = "mpmath"
+version = "1.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e0/47/dd32fa426cc72114383ac549964eecb20ecfd886d1e5ccf5340b55b02f57/mpmath-1.3.0.tar.gz", hash = "sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f", size = 508106, upload-time = "2023-03-07T16:47:11.061Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl", hash = "sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c", size = 536198, upload-time = "2023-03-07T16:47:09.197Z" },
+]
+
+[[package]]
+name = "networkx"
+version = "3.6.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/6a/51/63fe664f3908c97be9d2e4f1158eb633317598cfa6e1fc14af5383f17512/networkx-3.6.1.tar.gz", hash = "sha256:26b7c357accc0c8cde558ad486283728b65b6a95d85ee1cd66bafab4c8168509", size = 2517025, upload-time = "2025-12-08T17:02:39.908Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9e/c9/b2622292ea83fbb4ec318f5b9ab867d0a28ab43c5717bb85b0a5f6b3b0a4/networkx-3.6.1-py3-none-any.whl", hash = "sha256:d47fbf302e7d9cbbb9e2555a0d267983d2aa476bac30e90dfbe5669bd57f3762", size = 2068504, upload-time = "2025-12-08T17:02:38.159Z" },
+]
+
+[[package]]
+name = "numpy"
+version = "2.5.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/9a/80/db0b4559e57ec36362bedbb05530a87fafbcb6067708c946967a41d449e7/numpy-2.5.2.tar.gz", hash = "sha256:d482d171c406ae88c5b19cad3b6a1c4c5209f886ab74bc44c2c865c23f52d860", size = 20773161, upload-time = "2026-08-09T13:48:27.962Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e5/21/4947e0e9d6c9fc2e2ff15b8949049ee44f63adb9cacc729ab8793f97e712/numpy-2.5.2-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8ee9c4eeb8454b3660a8b53493563c3e121c2fc94fbd72b848ef814ed7b676a9", size = 15612696, upload-time = "2026-08-09T13:45:04.151Z" },
+    { url = "https://files.pythonhosted.org/packages/14/25/3f0be4c1b9fdf5dd5e708a6806978564d7c46a055c000496309ff2a2f8af/numpy-2.5.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:7999d4ddb0c4025018373fd787510d46e04c769467af22869707b3c1cfd459ab", size = 16974396, upload-time = "2026-08-09T13:45:11.316Z" },
+]
+
+[[package]]
+name = "omegaconf"
+version = "2.3.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "antlr4-python3-runtime" },
+    { name = "pyyaml" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ce/3d/e4b57b8d9008c6ebe0d5eff901f91d5700cf7bdb8c8863df817463a7fd5e/omegaconf-2.3.1.tar.gz", hash = "sha256:e5e7de64aeebeddaf8e6d3f7a783b32ac2a01c0fbd9c878012caecb891a1f42a", size = 3298472, upload-time = "2026-06-11T05:05:12.885Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a4/0e/152509871bf30df6fc38569f52a2db9b55dd41aae957adae50a053ac7778/omegaconf-2.3.1-py3-none-any.whl", hash = "sha256:3d701d14e9a8828f1edd28bb70b725908b34277cdd72cf7d6a83f94dadc6b6a0", size = 79502, upload-time = "2026-06-11T05:05:09.954Z" },
+]
+
+[[package]]
+name = "packaging"
+version = "26.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/7d/fa/3944b40b07da9ce895c0e6303a5ab7d53da063554f534556b134a54d6093/packaging-26.3.tar.gz", hash = "sha256:94edc256424af38762eb31306eed28beb9f0efc50a8837492c9d6fd6004aed79", size = 313412, upload-time = "2026-08-04T18:15:28.737Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/63/34/ba1c580383c9eada3711951fef0795c80b829a078d72188184bcab9dd527/packaging-26.3-py3-none-any.whl", hash = "sha256:d7193f7c8e4e93f444fde0262bf90af30e16fa0ad0ad44cb553c87339b23cd1c", size = 129956, upload-time = "2026-08-04T18:15:27.159Z" },
+]
+
+[[package]]
+name = "pandas"
+version = "3.0.5"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy" },
+    { name = "python-dateutil" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/be/4f/5f3422a2afec5ffc46308b79e53291365a93748b498ac2e58bead0197916/pandas-3.0.5.tar.gz", hash = "sha256:dca3734d6ab7c906e6730f0788b0a1dbb9f2467731f9711f77995c8e9d62d712", size = 4658219, upload-time = "2026-07-22T22:19:28.819Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e0/51/fac252f4a913ed5eabf3c11b880a9e8d5a6c10f0b2129d0462212d238b4d/pandas-3.0.5-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:73fa87b08a7ef706f8aafda39ddaccf2a99047bea62d8c88a0361bcafb2237bc", size = 10485458, upload-time = "2026-07-22T22:18:12.834Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/8c/1f29b5be8d3fc47dd7567eb167fabba2085879b31e0287ce7cba6d3d2ff4/pandas-3.0.5-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2a29c53d85ea98c5e792c59ef82ee9fbe6ca902c0d0adb6b23f45ef894cd7bf6", size = 11499501, upload-time = "2026-07-22T22:18:17.689Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/9a/ffbd852d58bd74a617fe2f8ee6a58a96982271ce41cf981eab22190b4a4b/pandas-3.0.5-cp312-cp312-pyemscripten_2024_0_wasm32.whl", hash = "sha256:b2acb4650527eec6822c3dadb2b771277b65e7dae7a267d4bccf65fd1bb3fbce", size = 7197652, upload-time = "2026-07-22T22:18:22.502Z" },
+]
+
+[[package]]
+name = "parameterized"
+version = "0.9.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ea/49/00c0c0cc24ff4266025a53e41336b79adaa5a4ebfad214f433d623f9865e/parameterized-0.9.0.tar.gz", hash = "sha256:7fc905272cefa4f364c1a3429cbbe9c0f98b793988efb5bf90aac80f08db09b1", size = 24351, upload-time = "2023-03-27T02:01:11.592Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/00/2f/804f58f0b856ab3bf21617cccf5b39206e6c4c94c2cd227bde125ea6105f/parameterized-0.9.0-py2.py3-none-any.whl", hash = "sha256:4e0758e3d41bea3bbd05ec14fc2c24736723f243b28d702081aef438c9372b1b", size = 20475, upload-time = "2023-03-27T02:01:09.31Z" },
+]
+
+[[package]]
+name = "pluggy"
+version = "1.6.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f9/e2/3e91f31a7d2b083fe6ef3fa267035b518369d9511ffab804f839851d2779/pluggy-1.6.0.tar.gz", hash = "sha256:7dcc130b76258d33b90f61b658791dede3486c3e6bfb003ee5c9bfb396dd22f3", size = 69412, upload-time = "2025-05-15T12:30:07.975Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/54/20/4d324d65cc6d9205fabedc306948156824eb9f0ee1633355a8f7ec5c66bf/pluggy-1.6.0-py3-none-any.whl", hash = "sha256:e920276dd6813095e9377c0bc5566d94c932c33b27a3e3945d8389c374dd4746", size = 20538, upload-time = "2025-05-15T12:30:06.134Z" },
+]
+
+[[package]]
+name = "pycparser"
+version = "3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/1b/7d/92392ff7815c21062bea51aa7b87d45576f649f16458d78b7cf94b9ab2e6/pycparser-3.0.tar.gz", hash = "sha256:600f49d217304a5902ac3c37e1281c9fe94e4d0489de643a9504c5cdfdfc6b29", size = 103492, upload-time = "2026-01-21T14:26:51.89Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0c/c3/44f3fbbfa403ea2a7c779186dc20772604442dde72947e7d01069cbe98e3/pycparser-3.0-py3-none-any.whl", hash = "sha256:b727414169a36b7d524c1c3e31839a521725078d7b2ff038656844266160a992", size = 48172, upload-time = "2026-01-21T14:26:50.693Z" },
+]
+
+[[package]]
+name = "pygments"
+version = "2.21.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/49/2e/ced460408999b33da6b31b0021b0f37d329e202d4169aeb164493778f25b/pygments-2.21.0.tar.gz", hash = "sha256:610ca751c9bc2492b38eb9a38a7fbc93edbbb2d7182edaf34e66ae493dee5c8c", size = 5005329, upload-time = "2026-08-17T08:02:48.824Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/71/46/17f022dd3e953bf20a04a028a21ec746d942f8d2af30fa0f124fa0e6a684/pygments-2.21.0-py3-none-any.whl", hash = "sha256:2363c69b61c4a97c838da3b130dcd6468f4848992b21a82f2a63ec34377137d9", size = 1250147, upload-time = "2026-08-17T08:02:44.912Z" },
+]
+
+[[package]]
+name = "pytest"
+version = "8.4.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "iniconfig" },
+    { name = "packaging" },
+    { name = "pluggy" },
+    { name = "pygments" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a3/5c/00a0e072241553e1a7496d638deababa67c5058571567b92a7eaa258397c/pytest-8.4.2.tar.gz", hash = "sha256:86c0d0b93306b961d58d62a4db4879f27fe25513d4b969df351abdddb3c30e01", size = 1519618, upload-time = "2025-09-04T14:34:22.711Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a8/a4/20da314d277121d6534b3a980b29035dcd51e6744bd79075a6ce8fa4eb8d/pytest-8.4.2-py3-none-any.whl", hash = "sha256:872f880de3fc3a5bdc88a11b39c9710c3497a547cfa9320bc3c5e62fbf272e79", size = 365750, upload-time = "2025-09-04T14:34:20.226Z" },
+]
+
+[[package]]
+name = "pytest-json-report"
+version = "1.5.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pytest" },
+    { name = "pytest-metadata" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/4f/d3/765dae9712fcd68d820338908c1337e077d5fdadccd5cacf95b9b0bea278/pytest-json-report-1.5.0.tar.gz", hash = "sha256:2dde3c647851a19b5f3700729e8310a6e66efb2077d674f27ddea3d34dc615de", size = 21241, upload-time = "2022-03-15T21:03:10.2Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/81/35/d07400c715bf8a88aa0c1ee9c9eb6050ca7fe5b39981f0eea773feeb0681/pytest_json_report-1.5.0-py3-none-any.whl", hash = "sha256:9897b68c910b12a2e48dd849f9a284b2c79a732a8a9cb398452ddd23d3c8c325", size = 13222, upload-time = "2022-03-15T21:03:08.65Z" },
+]
+
+[[package]]
+name = "pytest-metadata"
+version = "3.1.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pytest" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a6/85/8c969f8bec4e559f8f2b958a15229a35495f5b4ce499f6b865eac54b878d/pytest_metadata-3.1.1.tar.gz", hash = "sha256:d2a29b0355fbc03f168aa96d41ff88b1a3b44a3b02acbe491801c98a048017c8", size = 9952, upload-time = "2024-02-12T19:38:44.887Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3e/43/7e7b2ec865caa92f67b8f0e9231a798d102724ca4c0e1f414316be1c1ef2/pytest_metadata-3.1.1-py3-none-any.whl", hash = "sha256:c8e0844db684ee1c798cfa38908d20d67d0463ecb6137c72e91f418558dd5f4b", size = 11428, upload-time = "2024-02-12T19:38:42.531Z" },
+]
+
+[[package]]
+name = "pytest-rerunfailures"
+version = "15.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "packaging" },
+    { name = "pytest" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a0/78/e6e358545537a8e82c4dc91e72ec0d6f80546a3786dd27c76b06ca09db77/pytest_rerunfailures-15.1.tar.gz", hash = "sha256:c6040368abd7b8138c5b67288be17d6e5611b7368755ce0465dda0362c8ece80", size = 26981, upload-time = "2025-05-08T06:36:33.483Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f3/30/11d836ff01c938969efa319b4ebe2374ed79d28043a12bfc908577aab9f3/pytest_rerunfailures-15.1-py3-none-any.whl", hash = "sha256:f674c3594845aba8b23c78e99b1ff8068556cc6a8b277f728071fdc4f4b0b355", size = 13274, upload-time = "2025-05-08T06:36:32.029Z" },
+]
+
+[[package]]
+name = "pytest-xdist"
+version = "3.8.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "execnet" },
+    { name = "pytest" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/78/b4/439b179d1ff526791eb921115fca8e44e596a13efeda518b9d845a619450/pytest_xdist-3.8.0.tar.gz", hash = "sha256:7e578125ec9bc6050861aa93f2d59f1d8d085595d6551c2c90b6f4fad8d3a9f1", size = 88069, upload-time = "2025-07-01T13:30:59.346Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ca/31/d4e37e9e550c2b92a9cbc2e4d0b7420a27224968580b5a447f420847c975/pytest_xdist-3.8.0-py3-none-any.whl", hash = "sha256:202ca578cfeb7370784a8c33d6d05bc6e13b4f25b5053c30a152269fd10f0b88", size = 46396, upload-time = "2025-07-01T13:30:56.632Z" },
+]
+
+[[package]]
+name = "python-dateutil"
+version = "2.9.0.post0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "six" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/66/c0/0c8b6ad9f17a802ee498c46e004a0eb49bc148f2fd230864601a86dcf6db/python-dateutil-2.9.0.post0.tar.gz", hash = "sha256:37dd54208da7e1cd875388217d5e00ebd4179249f90fb72437e91a35459a0ad3", size = 342432, upload-time = "2024-03-01T18:36:20.211Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl", hash = "sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427", size = 229892, upload-time = "2024-03-01T18:36:18.57Z" },
+]
+
+[[package]]
+name = "pytorch-tokenizers"
+version = "1.4.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "sentencepiece" },
+    { name = "tiktoken" },
+    { name = "tokenizers" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d6/cf/69d6d3c6fad65f0f1c63a8f0f0db85009a9db4e25fa50d573ead4e527cbb/pytorch_tokenizers-1.4.1-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:84ee954611ec869f875dd23618c66c15b718c2a4eed639fefb72e7806091ff76", size = 1451907, upload-time = "2026-08-05T19:37:41.622Z" },
+]
+
+[[package]]
+name = "pyyaml"
+version = "6.0.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/05/8e/961c0007c59b8dd7729d542c61a4d537767a59645b82a0b521206e1e25c2/pyyaml-6.0.3.tar.gz", hash = "sha256:d76623373421df22fb4cf8817020cbb7ef15c725b9d5e45f17e189bfc384190f", size = 130960, upload-time = "2025-09-25T21:33:16.546Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ed/23/7a778b6bd0b9a8039df8b1b1d80e2e2ad78aa04171592c8a5c43a56a6af4/pyyaml-6.0.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9149cad251584d5fb4981be1ecde53a1ca46c891a79788c0df828d2f166bda28", size = 775116, upload-time = "2025-09-25T21:32:13.652Z" },
+    { url = "https://files.pythonhosted.org/packages/05/c0/b3be26a015601b822b97d9149ff8cb5ead58c66f981e04fedf4e762f4bd4/pyyaml-6.0.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:8dc52c23056b9ddd46818a57b78404882310fb473d63f17b07d5c40421e47f8e", size = 761089, upload-time = "2025-09-25T21:32:17.56Z" },
+]
+
+[[package]]
+name = "regex"
+version = "2026.7.19"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/20/98/04b13f1ddfb63158025291c02e03eb42fbb7acb51d091d541050eb4e35e8/regex-2026.7.19.tar.gz", hash = "sha256:7e77b324909c1617cbb4c668677e2c6ae13f44d7c1de0d4f15f2e3c10f3315b5", size = 416440, upload-time = "2026-07-19T00:19:48.923Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/22/85/102a81b218298957d4ea7d2f084fae537a71add9d6ff93c8e67284c5f45e/regex-2026.7.19-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:93db40c8de0815baab96a06e08a984bac71f989d13bab789e382158c5d426797", size = 796708, upload-time = "2026-07-19T00:17:11.542Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/69/03c9b3f058d66403e0ca2c938696e81d51cd4c6d47ec5265f02f96948d9a/regex-2026.7.19-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:1d793a7988e04fcb1e2e135567443d82173225d657419ec09414a9b5a145b986", size = 784338, upload-time = "2026-07-19T00:17:21.057Z" },
+]
+
+[[package]]
+name = "requests"
+version = "2.34.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "certifi" },
+    { name = "charset-normalizer" },
+    { name = "idna" },
+    { name = "urllib3" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ac/c3/e2a2b89f2d3e2179abd6d00ebd70bff6273f37fb3e0cc209f48b39d00cbf/requests-2.34.2.tar.gz", hash = "sha256:f288924cae4e29463698d6d60bc6a4da69c89185ad1e0bcc4104f584e960b9ed", size = 142856, upload-time = "2026-05-14T19:25:27.735Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a0/f4/c67b0b3f1b9245e8d266f0f112c500d50e5b4e83cb6f3b71b6528104182a/requests-2.34.2-py3-none-any.whl", hash = "sha256:2a0d60c172f83ac6ab31e4554906c0f3b3588d37b5cb939b1c061f4907e278e0", size = 73075, upload-time = "2026-05-14T19:25:26.443Z" },
+]
+
+[[package]]
+name = "ruamel-yaml"
+version = "0.19.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/c7/3b/ebda527b56beb90cb7652cb1c7e4f91f48649fbcd8d2eb2fb6e77cd3329b/ruamel_yaml-0.19.1.tar.gz", hash = "sha256:53eb66cd27849eff968ebf8f0bf61f46cdac2da1d1f3576dd4ccee9b25c31993", size = 142709, upload-time = "2026-01-02T16:50:31.84Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b8/0c/51f6841f1d84f404f92463fc2b1ba0da357ca1e3db6b7fbda26956c3b82a/ruamel_yaml-0.19.1-py3-none-any.whl", hash = "sha256:27592957fedf6e0b62f281e96effd28043345e0e66001f97683aa9a40c667c93", size = 118102, upload-time = "2026-01-02T16:50:29.201Z" },
+]
+
+[[package]]
+name = "safetensors"
+version = "0.8.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/45/06/f955dbbb1859e3bd23c8ac6141af5106e7ad5fedec4a3a6e3d60f94b7001/safetensors-0.8.0.tar.gz", hash = "sha256:fabaf3e0f18a6618d9b36560682562157f77c2b71fcffc7b432be2baed9d753d", size = 325846, upload-time = "2026-06-09T07:52:25.563Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/09/7d/65a7de0af421317bb36a067241e4235fff194eed60b961ed6d3f59a3fc60/safetensors-0.8.0-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7a46e5ff292c356d6991e60942ba7f79817682d3a2cef0702136448cb9c4d235", size = 502844, upload-time = "2026-06-09T07:52:07.624Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/43/bf38443278eab4b1be1fce2931e2b012ad9cb7df52ada751d0aab8f7659a/safetensors-0.8.0-cp310-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:87eec7ffed2b809f05a398a8becb7d013f19f7837cd15d9748580d6cf30dbaf4", size = 678670, upload-time = "2026-06-09T07:52:20.032Z" },
+]
+
+[[package]]
+name = "scikit-learn"
+version = "1.7.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "joblib" },
+    { name = "numpy" },
+    { name = "scipy" },
+    { name = "threadpoolctl" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/41/84/5f4af978fff619706b8961accac84780a6d298d82a8873446f72edb4ead0/scikit_learn-1.7.1.tar.gz", hash = "sha256:24b3f1e976a4665aa74ee0fcaac2b8fccc6ae77c8e07ab25da3ba6d3292b9802", size = 7190445, upload-time = "2025-07-18T08:01:54.5Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ad/09/a2aa0b4e644e5c4ede7006748f24e72863ba2ae71897fecfd832afea01b4/scikit_learn-1.7.1-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e3142f0abe1ad1d1c31a2ae987621e41f6b578144a911ff4ac94781a583adad7", size = 9290988, upload-time = "2025-07-18T08:01:28.938Z" },
+]
+
+[[package]]
+name = "scipy"
+version = "1.18.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/7e/74/66de6258867beb2ef08f35f9f2ac017a52cacd5081714d239ff1a442d458/scipy-1.18.1.tar.gz", hash = "sha256:52c4b7422442aba924d03ad4019852b08a92e64ea187b933135687bfe2747307", size = 30781235, upload-time = "2026-08-21T23:28:50.599Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/75/9a/2e71719f31eaefe0e3a1706c4a1ded94e664bfd95ffca2b219a671faee01/scipy-1.18.1-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3c085faa2cfa879c5141df483f836f4d691045a078224a670fa570fa01612d89", size = 34025113, upload-time = "2026-08-21T23:24:02.209Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/af/c5538be1792f7034c12c7db6ee67cace58253c7b87b122d68253eaf5de89/scipy-1.18.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:c35d74ce0e193ff740c2f2be2ac913ddc232fe6c1ff40b26cfecb9c670c63314", size = 35639587, upload-time = "2026-08-21T23:24:13.05Z" },
+]
+
+[[package]]
+name = "sentencepiece"
+version = "0.2.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/cc/33/ea3cb3839607eb175da835244a798f797f478c5ddf0e8ecdf57ea85a4c70/sentencepiece-0.2.2.tar.gz", hash = "sha256:3d2b5e824b5622038dc7b490897efe05ebbbb9e7350fc142f3ecc8789ef9bdf6", size = 8218435, upload-time = "2026-07-12T08:39:34.701Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/19/90/cd798935668cff71d309d8ff10385844ecf216b1fe454f1993ed8bf2cb91/sentencepiece-0.2.2-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:cbce24284f51f71d10a42b7b9c964dcb9048b28f1c8e5db40bcbcb6f428cba6a", size = 1325200, upload-time = "2026-07-12T08:38:33.689Z" },
+]
+
+[[package]]
+name = "setuptools"
+version = "84.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/6d/44/f5da03a8ef95d369145c5bb53050e7877c9f3d312e128605fd9504829143/setuptools-84.0.0.tar.gz", hash = "sha256:f4695c21257f0d9b537ec2692c941d02ee143b7cc1276941349a546573b2ef73", size = 1168449, upload-time = "2026-08-08T18:27:58.365Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/95/9c/c510029fc6ef33a6275cd2c5d3cecd6613dfd6aa401d57c54f1c18852ccf/setuptools-84.0.0-py3-none-any.whl", hash = "sha256:51a52592b3b99e102b609654876bd65f19f999935166d1352678931132b0c670", size = 818216, upload-time = "2026-08-08T18:27:56.719Z" },
+]
+
+[[package]]
+name = "six"
+version = "1.17.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/94/e7/b2c673351809dca68a0e064b6af791aa332cf192da575fd474ed7d6f16a2/six-1.17.0.tar.gz", hash = "sha256:ff70335d468e7eb6ec65b95b99d3a2836546063f63acc5171de367e834932a81", size = 34031, upload-time = "2024-12-04T17:35:28.174Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b7/ce/149a00dd41f10bc29e5921b496af8b574d8413afcd5e30dfa0ed46c2cc5e/six-1.17.0-py2.py3-none-any.whl", hash = "sha256:4721f391ed90541fddacab5acf947aa0d3dc7d27b2e1e8eda2be8970586c3274", size = 11050, upload-time = "2024-12-04T17:35:26.475Z" },
+]
+
+[[package]]
+name = "sortedcontainers"
+version = "2.4.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e8/c4/ba2f8066cceb6f23394729afe52f3bf7adec04bf9ed2c820b39e19299111/sortedcontainers-2.4.0.tar.gz", hash = "sha256:25caa5a06cc30b6b83d11423433f65d1f9d76c4c6a0c90e3379eaa43b9bfdb88", size = 30594, upload-time = "2021-05-16T22:03:42.897Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/32/46/9cb0e58b2deb7f82b84065f37f3bffeb12413f947f9388e4cac22c4621ce/sortedcontainers-2.4.0-py2.py3-none-any.whl", hash = "sha256:a163dcaede0f1c021485e957a39245190e74249897e2ae4b2aa38595db237ee0", size = 29575, upload-time = "2021-05-16T22:03:41.177Z" },
+]
+
+[[package]]
+name = "soundfile"
+version = "0.13.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "cffi" },
+    { name = "numpy" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/e1/41/9b873a8c055582859b239be17902a85339bec6a30ad162f98c9b0288a2cc/soundfile-0.13.1.tar.gz", hash = "sha256:b2c68dab1e30297317080a5b43df57e302584c49e2942defdde0acccc53f0e5b", size = 46156, upload-time = "2025-01-25T09:17:04.831Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/64/28/e2a36573ccbcf3d57c00626a21fe51989380636e821b341d36ccca0c1c3a/soundfile-0.13.1-py2.py3-none-any.whl", hash = "sha256:a23c717560da2cf4c7b5ae1142514e0fd82d6bbd9dfc93a50423447142f2c445", size = 25751, upload-time = "2025-01-25T09:16:44.235Z" },
+    { url = "https://files.pythonhosted.org/packages/58/ae/c0e4a53d77cf6e9a04179535766b3321b0b9ced5f70522e4caf9329f0046/soundfile-0.13.1-py2.py3-none-manylinux_2_28_aarch64.whl", hash = "sha256:9c9e855f5a4d06ce4213f31918653ab7de0c5a8d8107cd2427e44b42df547deb", size = 1235729, upload-time = "2025-01-25T09:16:53.018Z" },
+]
+
+[[package]]
+name = "sympy"
+version = "1.14.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "mpmath" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/83/d3/803453b36afefb7c2bb238361cd4ae6125a569b4db67cd9e79846ba2d68c/sympy-1.14.0.tar.gz", hash = "sha256:d3d3fe8df1e5a0b42f0e7bdf50541697dbe7d23746e894990c030e2b05e72517", size = 7793921, upload-time = "2025-04-27T18:05:01.611Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a2/09/77d55d46fd61b4a135c444fc97158ef34a095e5681d0a6c10b75bf356191/sympy-1.14.0-py3-none-any.whl", hash = "sha256:e091cc3e99d2141a0ba2847328f5479b05d94a6635cb96148ccb3f34671bd8f5", size = 6299353, upload-time = "2025-04-27T18:04:59.103Z" },
+]
+
+[[package]]
+name = "tabulate"
+version = "0.10.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/46/58/8c37dea7bbf769b20d58e7ace7e5edfe65b849442b00ffcdd56be88697c6/tabulate-0.10.0.tar.gz", hash = "sha256:e2cfde8f79420f6deeffdeda9aaec3b6bc5abce947655d17ac662b126e48a60d", size = 91754, upload-time = "2026-03-04T18:55:34.402Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/99/55/db07de81b5c630da5cbf5c7df646580ca26dfaefa593667fc6f2fe016d2e/tabulate-0.10.0-py3-none-any.whl", hash = "sha256:f0b0622e567335c8fabaaa659f1b33bcb6ddfe2e496071b743aa113f8774f2d3", size = 39814, upload-time = "2026-03-04T18:55:31.284Z" },
+]
+
+[[package]]
+name = "threadpoolctl"
+version = "3.6.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/b7/4d/08c89e34946fce2aec4fbb45c9016efd5f4d7f24af8e5d93296e935631d8/threadpoolctl-3.6.0.tar.gz", hash = "sha256:8ab8b4aa3491d812b623328249fab5302a68d2d71745c8a4c719a2fcaba9f44e", size = 21274, upload-time = "2025-03-13T13:49:23.031Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/32/d5/f9a850d79b0851d1d4ef6456097579a9005b31fea68726a4ae5f2d82ddd9/threadpoolctl-3.6.0-py3-none-any.whl", hash = "sha256:43a0b8fd5a2928500110039e43a5eed8480b918967083ea48dc3ab9f13c4a7fb", size = 18638, upload-time = "2025-03-13T13:49:21.846Z" },
+]
+
+[[package]]
+name = "tiktoken"
+version = "0.14.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "regex" },
+    { name = "requests" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/66/62/167a842aa0429d45f5e797354fd4343a96f6043d67d0513c675c7b8d36e6/tiktoken-0.14.0.tar.gz", hash = "sha256:231dec90efcdccf1b565a1416107736f1e09b1a08fe736ef9d6363e626d03874", size = 38898, upload-time = "2026-08-17T19:49:49.514Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0b/35/e9f47647c9e163bd1de30fe1a491669b7248cfc67b7404c35c009a701e1a/tiktoken-0.14.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:7db45b98e94adf4173a5cd7422b150999a7ee11ff847783a14f6e1b80cc38cb6", size = 1186355, upload-time = "2026-08-17T19:48:51.93Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/9c/7035b0bcfaa68d1ee4803fc5be5214ad865669b05bd20e7105ae8a18afc6/tiktoken-0.14.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:b950248272f1b303dc32986396e2dccfa10cf6d1e83ec8f0bba1776660305482", size = 1250635, upload-time = "2026-08-17T19:48:54.392Z" },
+]
+
+[[package]]
+name = "tokenizers"
+version = "0.22.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "huggingface-hub" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/73/6f/f80cfef4a312e1fb34baf7d85c72d4411afde10978d4657f8cdd811d3ccc/tokenizers-0.22.2.tar.gz", hash = "sha256:473b83b915e547aa366d1eee11806deaf419e17be16310ac0a14077f1e28f917", size = 372115, upload-time = "2026-01-05T10:45:15.988Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d6/84/7990e799f1309a8b87af6b948f31edaa12a3ed22d11b352eaf4f4b2e5753/tokenizers-0.22.2-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2249487018adec45d6e3554c71d46eb39fa8ea67156c640f7513eb26f318cec7", size = 3290736, upload-time = "2026-01-05T10:40:32.165Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/28/5f9f5a4cc211b69e89420980e483831bcc29dade307955cc9dc858a40f01/tokenizers-0.22.2-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:29c30b83d8dcd061078b05ae0cb94d3c710555fbb44861139f9f83dcca3dc3e4", size = 9478245, upload-time = "2026-01-05T10:41:04.053Z" },
+]
+
+[[package]]
+name = "torch"
+version = "2.10.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "filelock" },
+    { name = "fsspec" },
+    { name = "jinja2" },
+    { name = "networkx" },
+    { name = "setuptools" },
+    { name = "sympy" },
+    { name = "typing-extensions" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/cc/af/758e242e9102e9988969b5e621d41f36b8f258bb4a099109b7a4b4b50ea4/torch-2.10.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:5fd4117d89ffd47e3dcc71e71a22efac24828ad781c7e46aaaf56bf7f2796acf", size = 145996088, upload-time = "2026-01-21T16:24:44.171Z" },
+]
+
+[[package]]
+name = "torchao"
+version = "0.15.0"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f6/3b/6b9d5618720f63dbc2e2509cd6b57aae9c0d61b738d1d2172f4d5d9efaab/torchao-0.15.0-py3-none-any.whl", hash = "sha256:3f3812676048ef8a2a0e9d492d12d8971ba7a7ebb16f54aa56f690414e130d2c", size = 1080679, upload-time = "2025-12-18T23:14:43.807Z" },
+]
+
+[[package]]
+name = "tqdm"
+version = "4.70.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/21/3b/6c24bec5be5e743ffd99576daa5cc077722fc7d5bbc00bd133fa0c698dc6/tqdm-4.70.0.tar.gz", hash = "sha256:55b0b0dbd97462d06ebee91e4dac24ed4d4702be82b24f07e6c1d27e08cea220", size = 795438, upload-time = "2026-07-27T11:33:15.271Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f9/1c/01bfd571a64e7f270e6bab5e33777debe0edc56759233ce84f27dec92d14/tqdm-4.70.0-py3-none-any.whl", hash = "sha256:7f585706bfddbdebf89daac705b2dfcc16890130727d3197ca62c732b4310953", size = 80184, upload-time = "2026-07-27T11:33:13.167Z" },
+]
+
+[[package]]
+name = "transformers"
+version = "4.57.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "filelock" },
+    { name = "huggingface-hub" },
+    { name = "numpy" },
+    { name = "packaging" },
+    { name = "pyyaml" },
+    { name = "regex" },
+    { name = "requests" },
+    { name = "safetensors" },
+    { name = "tokenizers" },
+    { name = "tqdm" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/d6/68/a39307bcc4116a30b2106f2e689130a48de8bd8a1e635b5e1030e46fcd9e/transformers-4.57.1.tar.gz", hash = "sha256:f06c837959196c75039809636cd964b959f6604b75b8eeec6fdfc0440b89cc55", size = 10142511, upload-time = "2025-10-14T15:39:26.18Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/71/d3/c16c3b3cf7655a67db1144da94b021c200ac1303f82428f2beef6c2e72bb/transformers-4.57.1-py3-none-any.whl", hash = "sha256:b10d05da8fa67dc41644dbbf9bc45a44cb86ae33da6f9295f5fbf5b7890bd267", size = 11990925, upload-time = "2025-10-14T15:39:23.085Z" },
+]
+
+[[package]]
+name = "typing-extensions"
+version = "4.16.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f6/cc/6253133b5bb138fc3306cebfbda2c520f545d36b5be2c7255cc528bb45d6/typing_extensions-4.16.0.tar.gz", hash = "sha256:dc983d19a509c94dba722ee6abd33940f7c05a89e243c47e907eb4db6f1a43e5", size = 113555, upload-time = "2026-07-02T08:40:05.92Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/49/d3/b8441a820a491ddfc024b0b0cf0393375b75ea13866d9c66727e54c2fc80/typing_extensions-4.16.0-py3-none-any.whl", hash = "sha256:481caa481374e813c1b176ada14e97f1f67a4539ce9cfeb3f350d78d6370c2e8", size = 45571, upload-time = "2026-07-02T08:40:04.659Z" },
+]
+
+[[package]]
+name = "urllib3"
+version = "2.7.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/53/0c/06f8b233b8fd13b9e5ee11424ef85419ba0d8ba0b3138bf360be2ff56953/urllib3-2.7.0.tar.gz", hash = "sha256:231e0ec3b63ceb14667c67be60f2f2c40a518cb38b03af60abc813da26505f4c", size = 433602, upload-time = "2026-05-07T16:13:18.596Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/7f/3e/5db95bcf282c52709639744ca2a8b149baccf648e39c8cc87553df9eae0c/urllib3-2.7.0-py3-none-any.whl", hash = "sha256:9fb4c81ebbb1ce9531cce37674bbc6f1360472bc18ca9a553ede278ef7276897", size = 131087, upload-time = "2026-05-07T16:13:17.151Z" },
+]
+
+[[package]]
+name = "whisper-small-int8-xnnpack-executorch-runtime"
+version = "0.1.0"
+source = { virtual = "." }
+dependencies = [
+    { name = "executorch" },
+    { name = "numpy" },
+    { name = "soundfile" },
+    { name = "torch" },
+    { name = "transformers" },
+]
+
+[package.metadata]
+requires-dist = [
+    { name = "executorch", specifier = "==1.1.0" },
+    { name = "numpy", specifier = "==2.5.2" },
+    { name = "soundfile", specifier = "==0.13.1" },
+    { name = "torch", specifier = "==2.10.0" },
+    { name = "transformers", specifier = "==4.57.1" },
+]