> This page is for version 26.07 · v1.3.0.
> For other versions, use one of these documentation indexes:
> - Latest · v1.4.0 (26.09) (default): https://docs.nvidia.com/nemo/curator/latest/llms.txt
> - Main · preview: https://docs.nvidia.com/nemo/curator/main/llms.txt
> - 26.09 · v1.4.0: https://docs.nvidia.com/nemo/curator/v26.09/llms.txt
> - 26.07 · v1.3.0: https://docs.nvidia.com/nemo/curator/v26.07/llms.txt
> - 26.04 · v1.2.0: https://docs.nvidia.com/nemo/curator/v26.04/llms.txt
> - 26.02 · v1.1.0: https://docs.nvidia.com/nemo/curator/v26.02/llms.txt
> - 25.09 · v1.0.0: https://docs.nvidia.com/nemo/curator/v25.09/llms.txt

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/curator/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/curator/_mcp/server.

# nemo_curator.models.asr.faster_whisper

Faster-Whisper implementation of the shared ASR adapter.

## Module Contents

### Classes

| Name                                                                           | Description                                                         |
| ------------------------------------------------------------------------------ | ------------------------------------------------------------------- |
| [`FasterWhisperASR`](#nemo_curator-models-asr-faster_whisper-FasterWhisperASR) | Run Faster-Whisper on mono 16 kHz waveforms prepared by `ASRStage`. |

### Functions

| Name                                                                                         | Description |
| -------------------------------------------------------------------------------------------- | ----------- |
| [`_download_whisper_model`](#nemo_curator-models-asr-faster_whisper-_download_whisper_model) | -           |
| [`_faster_whisper_stack`](#nemo_curator-models-asr-faster_whisper-_faster_whisper_stack)     | -           |
| [`_whisper_model_class`](#nemo_curator-models-asr-faster_whisper-_whisper_model_class)       | -           |

### Data

[`_LANGUAGE_ALIASES`](#nemo_curator-models-asr-faster_whisper-_LANGUAGE_ALIASES)

[`_TARGET_SAMPLE_RATE`](#nemo_curator-models-asr-faster_whisper-_TARGET_SAMPLE_RATE)

### API

```python
class nemo_curator.models.asr.faster_whisper.FasterWhisperASR(
    model_id: str = 'large-v3',
    revision: str | None = None,
    compute_type: str = 'float16',
    beam_size: int = 5,
    vad_filter: bool = True,
    without_timestamps: bool = True,
    cpu_compute_type: str = 'int8'
)
```

Dataclass

Run Faster-Whisper on mono 16 kHz waveforms prepared by `ASRStage`.

**`_model`** `Any | None = field(default=None, init=False, repr=False)`

---

**`beam_size`** `int = 5`

---

**`compute_type`** `str = 'float16'`

---

**`cpu_compute_type`** `str = 'int8'`

---

**`model_id`** `str = 'large-v3'`

---

**`revision`** `str | None = None`

---

**`vad_filter`** `bool = True`

---

**`without_timestamps`** `bool = True`

---

```python
nemo_curator.models.asr.faster_whisper.FasterWhisperASR.__post_init__() -> None
```

```python
nemo_curator.models.asr.faster_whisper.FasterWhisperASR.download_weights_on_node() -> None
```

Populate Faster-Whisper's node-local cache without allocating a GPU.

```python
nemo_curator.models.asr.faster_whisper.FasterWhisperASR.load_model(
    num_gpus: int
) -> None
```

```python
nemo_curator.models.asr.faster_whisper.FasterWhisperASR.transcribe_batch(
    items: list[dict[str, typing.Any]]
) -> list[nemo_curator.models.asr.base.ASRResult]
```

```python
nemo_curator.models.asr.faster_whisper.FasterWhisperASR.unload_model() -> None
```

```python
nemo_curator.models.asr.faster_whisper._download_whisper_model(
    model_id: str,
    revision: str | None
) -> None
```

```python
nemo_curator.models.asr.faster_whisper._faster_whisper_stack() -> tuple[type, typing.Any]
```

```python
nemo_curator.models.asr.faster_whisper._whisper_model_class() -> type
```

```python
nemo_curator.models.asr.faster_whisper._LANGUAGE_ALIASES = {'fil': 'tl', 'in': 'id', 'iw': 'he', 'ji': 'yi', 'jv': 'jw', 'nb': 'no', 'tl': ...
```

```python
nemo_curator.models.asr.faster_whisper._TARGET_SAMPLE_RATE = 16000
```