---
language: 
- en
- zh
- de
- es
- ru
- ko
- fr
- ja
- pt
- tr
- pl
- ca
- nl
- ar
- sv
- it
- id
- hi
- fi
- vi
- he
- uk
- el
- ms
- cs
- ro
- da
- hu
- ta
- no
- th
- ur
- hr
- bg
- lt
- la
- mi
- ml
- cy
- sk
- te
- fa
- lv
- bn
- sr
- az
- sl
- kn
- et
- mk
- br
- eu
- is
- hy
- ne
- mn
- bs
- kk
- sq
- sw
- gl
- mr
- pa
- si
- km
- sn
- yo
- so
- af
- oc
- ka
- be
- tg
- sd
- gu
- am
- yi
- lo
- uz
- fo
- ht
- ps
- tk
- nn
- mt
- sa
- lb
- my
- bo
- tl
- mg
- as
- tt
- haw
- ln
- ha
- ba
- jw
- su
tags:
- audio
- automatic-speech-recognition
- hf-asr-leaderboard
pipeline_tag: automatic-speech-recognition
license: mit
license_link: https://choosealicense.com/licenses/mit/
base_model: openai/whisper-large-v3-turbo
---

# whisper-large-v3-turbo-fp16-ov
* Model creator: [OpenAI](https://huggingface.co/openai)
 * Original model: [whisper-large-v3-turbo](https://huggingface.co/openai/whisper-large-v3-turbo)

## Description
This is [whisper-large-v3-turbo](https://huggingface.co/openai/whisper-large-v3-turbo) model converted to the [OpenVINO™ IR](https://docs.openvino.ai/2025/documentation/openvino-ir-format.html) (Intermediate Representation) format with weights compressed to FP16.

Whisper large-v3-turbo is a finetuned version of a pruned [Whisper large-v3](https://huggingface.co/openai/whisper-large-v3) with the number of decoder layers reduced from 32 to 4, which makes the model significantly faster at the expense of a minor quality degradation.

## Compatibility

The provided OpenVINO™ IR model is compatible with:

* OpenVINO version 2026.1.0 and higher
* Optimum Intel 1.27.0 and higher


## Running Model Inference with [Optimum Intel](https://huggingface.co/docs/optimum/intel/index)

1. Install packages required for using [Optimum Intel](https://huggingface.co/docs/optimum/intel/index) integration with the OpenVINO backend:

```
pip install optimum[openvino] "datasets<4" librosa soundfile --extra-index-url https://download.pytorch.org/whl/cpu
```

2. Run model inference:

```
from datasets import load_dataset
from transformers import AutoProcessor
from optimum.intel.openvino import OVModelForSpeechSeq2Seq

model_id = "OpenVINO/whisper-large-v3-turbo-fp16-ov"
tokenizer = AutoProcessor.from_pretrained(model_id)
model = OVModelForSpeechSeq2Seq.from_pretrained(model_id)

dataset = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation", trust_remote_code=True)
sample = dataset[0]

input_features = tokenizer(
    sample["audio"]["array"],
    sampling_rate=sample["audio"]["sampling_rate"],
    return_tensors="pt",
).input_features

outputs = model.generate(input_features)
text = tokenizer.batch_decode(outputs)[0]
print(text)
```

## Running Model Inference with [OpenVINO GenAI](https://github.com/openvinotoolkit/openvino.genai)

1. Install packages required for using OpenVINO GenAI.
```
pip install huggingface_hub "datasets<4" librosa soundfile
pip install -U --pre --extra-index-url https://storage.openvinotoolkit.org/simple/wheels/nightly openvino openvino-tokenizers openvino-genai
```

2. Download model from HuggingFace Hub
   
```
import huggingface_hub as hf_hub

model_id = "OpenVINO/whisper-large-v3-turbo-fp16-ov"
model_path = "whisper-large-v3-turbo-fp16-ov"

hf_hub.snapshot_download(model_id, local_dir=model_path)

```

3. Run model inference:

```
import openvino_genai as ov_genai
import datasets

device = "CPU"
pipe = ov_genai.WhisperPipeline(model_path, device)

dataset = datasets.load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation", trust_remote_code=True)
sample = dataset[0]["audio"]["array"]
print(pipe.generate(sample))
```

More GenAI usage examples can be found in OpenVINO GenAI library [docs](https://github.com/openvinotoolkit/openvino.genai/blob/master/src/README.md) and [samples](https://github.com/openvinotoolkit/openvino.genai?tab=readme-ov-file#openvino-genai-samples)

## Running Model with OpenAI client and [OpenVINO Model Server](https://github.com/openvinotoolkit/model_server)

1a. Deploy model on Windows using [binary package](https://docs.openvino.ai/ovms_baremetal):
```bat
mkdir C:\models
ovms.exe --rest_port 8000 --source_model OpenVINO/whisper-large-v3-turbo-fp16-ov --model_repository_path C:\models
```

1b. Deploy model in a Docker container:
```bash
mkdir -p ${HOME}/models
export GPU_ARGS=$(if ls /dev/dri/render* >/dev/null 2>&1; then echo "--device /dev/dri --group-add $(stat -c '%g' /dev/dri/render* | head -n1)"; fi)
docker run ${GPU_ARGS} --rm --user $(id -u):$(id -g) -p 8000:8000 -v ${HOME}/models:/models openvino/model_server:latest-gpu --rest_port 8000 --model_repository_path /models --source_model OpenVINO/whisper-large-v3-turbo-fp16-ov
```

2. Install the client library:
```bash
pip install openai datasets soundfile
```

3. Run the client:
```python
import io

import soundfile as sf
from datasets import Audio, load_dataset
from openai import OpenAI


dataset = load_dataset(
    "hf-internal-testing/librispeech_asr_dummy",
    "clean",
    split="validation",
).cast_column("audio", Audio(decode=False))
audio_bytes = dataset[0]["audio"]["bytes"]

data, rate = sf.read(io.BytesIO(audio_bytes))
buffer = io.BytesIO()
sf.write(buffer, data, rate, format="WAV")

client = OpenAI(base_url="http://localhost:8000/v1", api_key="not_used")
for event in client.audio.transcriptions.create(
    model="OpenVINO/whisper-large-v3-turbo-fp16-ov",
    file=("sample.wav", buffer.getvalue()),
    language="en",
    stream=True,
):
    if getattr(event, "type", None) == "transcript.text.delta":
        print(event.delta, end="", flush=True)
    elif getattr(event, "type", None) == "transcript.text.done":
        print()
        break
```

## Limitations

Check the original model card for [original model card](https://huggingface.co/openai/whisper-large-v3-turbo) for limitations.

## Legal information

The original model is distributed under [MIT](https://choosealicense.com/licenses/mit/) license. More details can be found in [original model card](https://huggingface.co/openai/whisper-large-v3-turbo).

## Disclaimer

Intel is committed to respecting human rights and avoiding causing or contributing to adverse impacts on human rights. See [Intel’s Global Human Rights Principles](https://www.intel.com/content/dam/www/central-libraries/us/en/documents/policy-human-rights.pdf). Intel’s products and software are intended only to be used in applications that do not cause or contribute to adverse impacts on human rights.