> This page is for version Latest · v1.4.0 (26.09) (default).
> 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.nemo_asr

NeMo Framework ASR behind the shared `ASRAdapter` contract.

## Module Contents

### Classes

| Name                                                                 | Description                                                              |
| -------------------------------------------------------------------- | ------------------------------------------------------------------------ |
| [`NeMoASRAdapter`](#nemo_curator-models-asr-nemo_asr-NeMoASRAdapter) | Run a pretrained NeMo checkpoint using waveforms prepared by `ASRStage`. |

### Functions

| Name                                                                                                       | Description                                                            |
| ---------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------- |
| [`_extract_nemo_transcription_texts`](#nemo_curator-models-asr-nemo_asr-_extract_nemo_transcription_texts) | Extract text from the output shapes used by supported NeMo ASR models. |
| [`_nemo_asr_module`](#nemo_curator-models-asr-nemo_asr-_nemo_asr_module)                                   | -                                                                      |

### API

```python
class nemo_curator.models.asr.nemo_asr.NeMoASRAdapter(
    model_id: str = _DEFAULT_FASTCONFORMER_CTC_...,
    num_workers: int = 0,
    empty_audio_marks_skip: bool = True,
    verbose: bool = False,
    enable_local_attention: bool = False,
    local_attention_context_size: tuple[int, int] = (128, 128),
    use_cuda_graph_decoder: bool | None = None,
    refresh_cache: bool = False,
    strict: bool = True
)
```

Dataclass

Run a pretrained NeMo checkpoint using waveforms prepared by `ASRStage`.

**Parameters:**

**`model_id`** `str` — default: \_DEFAULT\_FASTCONFORMER\_CTC\_MODEL

Pretrained NeMo ASR checkpoint name.

---

**`num_workers`** `int` — default: 0

Data-loader workers used by NeMo's transcription call.

---

**`verbose`** `bool` — default: False

Forward NeMo transcription progress output.

---

**`enable_local_attention`** `bool` — default: False

Convert a compatible FastConformer checkpoint
from global to local attention after loading.

---

**`local_attention_context_size`** `tuple[int, int]` — default: (128, 128)

Left and right local-attention context.

---

**`use_cuda_graph_decoder`** `bool | None` — default: None

Override NeMo's RNNT CUDA-graph decoder. Leave
as `None` to preserve the checkpoint default. Set to `False`
on GPU/driver combinations that do not support NeMo's label-loop
CUDA graph implementation.

---

**`refresh_cache`** `bool` — default: False

Forward NeMo's checkpoint cache refresh flag.

---

**`strict`** `bool` — default: True

Forward NeMo's strict checkpoint loading flag.

---

**`_ATTENTION_CONTEXT_DIRECTIONS`** `int = 2`

---

**`_DEFAULT_FASTCONFORMER_CTC_MODEL`** `str = 'nvidia/stt_en_fastconformer_ctc_large'`

---

**`_DEFAULT_SAMPLE_RATE`** `int = 16000`

---

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

---

**`empty_audio_marks_skip`** `bool = True`

---

**`enable_local_attention`** `bool = False`

---

**`local_attention_context_size`** `tuple[int, int] = (128, 128)`

---

**`model_id`** `str = _DEFAULT_FASTCONFORMER_CTC_MODEL`

---

**`num_workers`** `int = 0`

---

**`refresh_cache`** `bool = False`

---

**`strict`** `bool = True`

---

**`use_cuda_graph_decoder`** `bool | None = None`

---

**`verbose`** `bool = False`

---

```python
nemo_curator.models.asr.nemo_asr.NeMoASRAdapter.__post_init__() -> None
```

```python
nemo_curator.models.asr.nemo_asr.NeMoASRAdapter._configure_local_attention(
    model: typing.Any
) -> None
```

```python
nemo_curator.models.asr.nemo_asr.NeMoASRAdapter._configure_rnnt_cuda_graph_decoder(
    model: typing.Any
) -> None
```

Override the CUDA-graph setting on a compatible NeMo RNNT decoder.

```python
nemo_curator.models.asr.nemo_asr.NeMoASRAdapter._load_checkpoint(
    device: typing.Any
) -> typing.Any
```

```python
nemo_curator.models.asr.nemo_asr.NeMoASRAdapter._transcribe_waveforms(
    waveforms: list[numpy.ndarray]
) -> list[str]
```

Run one NeMo call for the batch already bounded by `ASRStage`.

```python
nemo_curator.models.asr.nemo_asr.NeMoASRAdapter.download_weights_on_node() -> None
```

Download a pretrained checkpoint without allocating a GPU model.

```python
nemo_curator.models.asr.nemo_asr.NeMoASRAdapter.load_model(
    num_gpus: int
) -> None
```

Load one worker-local model on the device requested by `ASRStage`.

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

Transcribe one adapter call while preserving input order.

```python
nemo_curator.models.asr.nemo_asr.NeMoASRAdapter.unload_model() -> None
```

Release worker-local model and CUDA cache state.

```python
nemo_curator.models.asr.nemo_asr._extract_nemo_transcription_texts(
    outputs: object
) -> list[str]
```

Extract text from the output shapes used by supported NeMo ASR models.

```python
nemo_curator.models.asr.nemo_asr._nemo_asr_module() -> typing.Any
```