| | |
|---|
| | | Notes: Support recognition of single audio file, as well as file list in Kaldi-style wav.scp format: `wav_id wav_pat` |
|---|
| | | |
|---|
| | | ### Speech Recognition (Non-streaming) |
|---|
| | | #### SenseVoice |
|---|
| | | ```python |
|---|
| | | from funasr import AutoModel |
|---|
| | | from funasr.utils.postprocess_utils import rich_transcription_postprocess |
|---|
| | | |
|---|
| | | model_dir = "iic/SenseVoiceSmall" |
|---|
| | | |
|---|
| | | model = AutoModel( |
|---|
| | | model=model_dir, |
|---|
| | | vad_model="fsmn-vad", |
|---|
| | | vad_kwargs={"max_single_segment_time": 30000}, |
|---|
| | | device="cuda:0", |
|---|
| | | ) |
|---|
| | | |
|---|
| | | # en |
|---|
| | | res = model.generate( |
|---|
| | | input=f"{model.model_path}/example/en.mp3", |
|---|
| | | cache={}, |
|---|
| | | language="auto", # "zn", "en", "yue", "ja", "ko", "nospeech" |
|---|
| | | use_itn=True, |
|---|
| | | batch_size_s=60, |
|---|
| | | merge_vad=True, # |
|---|
| | | merge_length_s=15, |
|---|
| | | ) |
|---|
| | | text = rich_transcription_postprocess(res[0]["text"]) |
|---|
| | | print(text) |
|---|
| | | ``` |
|---|
| | | Parameter Descriptions: |
|---|
| | | - `model_dir`: The name of the model, or the model's path on the local disk. |
|---|
| | | - `trust_remote_code`: |
|---|
| | | - When set to `True`, this indicates that the model's code implementation should be loaded from the location specified by `remote_code`, which points to the exact code for the model (for example, `model.py` in the current directory). It supports absolute paths, relative paths, and network URLs. |
|---|
| | | - When set to `False`, this signifies that the model's code implementation is the integrated version within [FunASR](https://github.com/modelscope/FunASR). In this case, any modifications to `model.py` in the current directory will not take effect because the version loaded is the internal one from FunASR. For the model code, [click here to view](https://github.com/modelscope/FunASR/tree/main/funasr/models/sense_voice). |
|---|
| | | - `max_single_segment_time`: The maximum length of audio segments that the `vad_model` can cut, measured in milliseconds (ms). |
|---|
| | | - `use_itn`: Indicates whether the output should include punctuation and inverse text normalization. |
|---|
| | | - `batch_size_s`: Represents a dynamic batch size where the total duration of the audio in the batch is measured in seconds (s). |
|---|
| | | - `merge_vad`: Whether to concatenate short audio fragments cut by the vad model, with the merged length being `merge_length_s`, measured in seconds (s). |
|---|
| | | |
|---|
| | | #### Paraformer |
|---|
| | | ```python |
|---|
| | | from funasr import AutoModel |
|---|
| | | # paraformer-zh is a multi-functional asr model |
|---|
