Upload folder using huggingface_hub

README.md

@@ -0,0 +1,103 @@
+---
+license: cc-by-4.0
+language:
+  - tr
+pipeline_tag: text-to-speech
+library_name: pocket-tts
+datasets:
+  - serdarcaglar/turkish-tts-audiobooks
+base_model:
+  - kyutai/pocket-tts
+tags:
+  - pocket-tts
+  - tts
+  - turkish
+  - turkish-tts
+  - speech
+  - voice-cloning
+---
+# Pocket TTS Turkish — v0.1 Base
+
+Turkish 24-layer Pocket TTS teacher model based on Kyutai Pocket TTS.
+
+This is the **v0.1 base release**, focused on high-quality Turkish speech synthesis. It is intended both for direct Turkish TTS inference and as a base checkpoint for further fine-tuning and depth distillation.
+
+## Model
+
+* Architecture: Pocket TTS
+* Language: Turkish
+* Transformer depth: 24 layers
+* Tokenizer: Turkish SentencePiece BPE
+* Vocabulary size: 4,000
+* Audio sample rate: 24 kHz
+* Training checkpoint selected: 112,000 steps
+* Generation temperature: 0.3
+
+## Training data
+
+The model was trained primarily on:
+
+`serdarcaglar/turkish-tts-audiobooks`
+
+A leakage-controlled training/dev/test split and additional preprocessing, normalization, alignment validation and filtering were applied before training.
+
+## Usage
+
+With Pocket TTS:
+
+```bash
+uvx pocket-tts generate \
+  --config "hf://kaanhgunay/pocket-tts-tr/config.yaml@v0.1-base" \
+  --voice "./voice.wav" \
+  --text "Merhaba, bugün güzel bir gün." \
+  --output-path output.wav
+```
+
+Replace `HF_USERNAME` with the repository owner's Hugging Face username.
+
+## Strengths
+
+The model performs well on standard Turkish text, including:
+
+* fluent Turkish sentences
+* long-form sentences
+* Turkish morphology and long agglutinative words
+* numbers and dates when appropriately normalized
+* punctuation and natural pauses
+* voice-conditioned speech generation
+
+## Known limitations
+
+### Foreign-language words
+
+This release is primarily trained for Turkish.
+
+English and other foreign-language words may be pronounced according to Turkish orthographic and phonetic patterns. Code-switching is **not considered a supported capability in v0.1**.
+
+A future release is planned to improve Turkish-English code-switching while preserving Turkish quality.
+
+### Beginning-of-generation artifact
+
+A short transient or pronunciation artifact may occasionally be audible around the first syllable at the beginning of generated speech.
+
+This behavior is being investigated separately from the acoustic/language quality of the model.
+
+## Versioning
+
+* **v0.1-base** — Turkish 24L base teacher, 112k steps
+* Future releases may add code-switch improvements and distilled variants.
+
+## Attribution
+
+This model is derived from Kyutai Pocket TTS.
+
+Training data attribution:
+
+Serdar I. Çağlar, *Turkish TTS Audiobooks: a 2,724-hour Turkish read-speech corpus for text-to-speech*, Hugging Face, 2026.
+
+## License
+
+Model weights are released under CC BY 4.0.
+
+Users are responsible for ensuring that voice cloning and generated speech are used lawfully and with appropriate consent.
+

SHA256SUMS

@@ -0,0 +1,3 @@
+575d7a90e41de407340e8048e6c1b3d6573c36ab9054e1233d3cb0d6777eaf5f  model.safetensors
+c9ff291390865a78112e23b8ded6842300c6a88a58b0ef9694d6e8282f4ea490  tokenizer.model
+350c8145552415cfe092c3fa8f3181715b7e3c666ee7bab46b8480b36668a553  config.yaml

config.yaml

@@ -0,0 +1,55 @@
+weights_path: hf://kaanhgunay/pocket-tts-tr/model.safetensors@v0.1-base
+weights_path_without_voice_cloning: null
+default_temperature: 0.3
+flow_lm:
+  insert_bos_before_voice: true
+  dtype: float32
+  flow:
+    depth: 6
+    dim: 512
+  transformer:
+    d_model: 1024
+    hidden_scale: 4
+    max_period: 10000
+    num_heads: 16
+    num_layers: 24
+  lookup_table:
+    dim: 1024
+    n_bins: 4000
+    tokenizer: sentencepiece
+    tokenizer_path: hf://kaanhgunay/pocket-tts-tr/tokenizer.model@v0.1-base
+mimi:
+  dtype: float32
+  sample_rate: 24000
+  inner_dim: 32
+  outer_dim: 512
+  channels: 1
+  frame_rate: 12.5
+  seanet:
+    dimension: 512
+    channels: 1
+    n_filters: 64
+    n_residual_layers: 1
+    ratios:
+    - 6
+    - 5
+    - 4
+    kernel_size: 7
+    residual_kernel_size: 3
+    last_kernel_size: 3
+    dilation_base: 2
+    pad_mode: constant
+    compress: 2
+  transformer:
+    d_model: 512
+    num_heads: 8
+    num_layers: 2
+    layer_scale: 0.01
+    context: 250
+    dim_feedforward: 2048
+    input_dimension: 512
+    output_dimensions:
+    - 512
+  quantizer:
+    dimension: 32
+    output_dimension: 512

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:575d7a90e41de407340e8048e6c1b3d6573c36ab9054e1233d3cb0d6777eaf5f
+size 1344315200

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9ff291390865a78112e23b8ded6842300c6a88a58b0ef9694d6e8282f4ea490
+size 302928

training/tr_24l_teacher.yaml

@@ -0,0 +1,71 @@
+model_config: pocket_tts/config/english_2026-04_24l.yaml
+
+model_overrides:
+  flow_lm.lookup_table.tokenizer_path: data/tr/tokenizer_full.model
+
+start_from_pretrained: true
+reset_text_embedding: true
+
+data:
+  train_jsonl: data/tr/alignments/train_final_aligned_clean.jsonl
+  valid_jsonl: data/tr/alignments/dev_final_aligned_clean.jsonl
+  max_duration_sec: 30.0
+  max_voice_prompt_sec: 5.0
+  shuffle: true
+  loader_procs: 3
+
+flow:
+  type: lsd
+  kwargs: {}
+
+flow_batch_multiplier: 4
+eos_loss_weight: 0.1
+
+text_dropout: 0.2
+voice_dropout: 0.2
+
+optim:
+  lr: 2e-4
+  weight_decay: 0.1
+  betas:
+    - 0.9
+    - 0.95
+  eps: 1e-8
+  max_norm: 1.0
+  warmup_steps: 1000
+  schedule: constant
+  lr_min_ratio: 0.0
+
+run_dir: runs/tr_24l_teacher
+
+# RTX 5090 / 32 GB
+# effective batch = 16 * 4 = 64
+batch_size: 16
+grad_accum_steps: 4
+
+max_steps: 250000
+seed: 42
+
+stats_ema_steps: 1000
+stats_ema_decay: 0.999
+ema_decay: 0.999
+
+compile: true
+
+log_freq: 50
+valid_freq: 2500
+num_valid_batches: 50
+
+ckpt_freq: 1000
+num_ckpt_keep: 6
+
+sample_freq: 2500
+sample_temp: 0.3
+sample_cfg_coef: 2.0
+
+sample_sentences:
+  - "Bugün hava oldukça güzel ve insanlar parkta yürüyüş yapıyor."
+  - "Türkçe konuşma sentezi doğal, anlaşılır ve akıcı bir ses üretmelidir."
+  - "Otuz sekiz yolcu sabah erkenden tren istasyonunda buluştu."
+  - "İnsan bazen geçmişi düşünür, sonra sessizce gülümseyerek yoluna devam eder."
+  - "Çocuk küçük kırmızı topu bahçenin öbür ucuna doğru hızla yuvarladı."