admin-valentin/WhisperLiveKit
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

default diarization backend in now sortformer

Browse source
This commit is contained in:
Quentin Fuxa 2025-08-24 18:32:01 +02:00
parent ce781831ee
commit 5258305745
3 changed files with 3 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
README.md
View file

@ -186,7 +186,7 @@ The package includes an HTML/JavaScript implementation [here](https://github.com
| Diarization options | Description | Default | | Diarization options | Description | Default |
|-----------|-------------|---------| |-----------|-------------|---------|
| `--diarization` | Enable speaker identification | `False` | | `--diarization` | Enable speaker identification | `False` |
| `--diarization-backend` | `diart` or `sortformer` | `diart` | | `--diarization-backend` | `diart` or `sortformer` | `sortformer` |
| `--punctuation-split` | Use punctuation to improve speaker boundaries | `True` | | `--punctuation-split` | Use punctuation to improve speaker boundaries | `True` |
| `--segmentation-model` | Hugging Face model ID for Diart segmentation model. [Available models](https://github.com/juanmc2005/diart/tree/main?tab=readme-ov-file#pre-trained-models) | `pyannote/segmentation-3.0` | | `--segmentation-model` | Hugging Face model ID for Diart segmentation model. [Available models](https://github.com/juanmc2005/diart/tree/main?tab=readme-ov-file#pre-trained-models) | `pyannote/segmentation-3.0` |
| `--embedding-model` | Hugging Face model ID for Diart embedding model. [Available models](https://github.com/juanmc2005/diart/tree/main?tab=readme-ov-file#pre-trained-models) | `speechbrain/spkrec-ecapa-voxceleb` | | `--embedding-model` | Hugging Face model ID for Diart embedding model. [Available models](https://github.com/juanmc2005/diart/tree/main?tab=readme-ov-file#pre-trained-models) | `speechbrain/spkrec-ecapa-voxceleb` |

2
whisperlivekit/core.py
View file

@ -57,7 +57,7 @@ class TranscriptionEngine:
"static_init_prompt": None, "static_init_prompt": None,
"max_context_tokens": None, "max_context_tokens": None,
"model_path": './base.pt', "model_path": './base.pt',
"diarization_backend": "diart", "diarization_backend": "sortformer",
# diart params: # diart params:
"segmentation_model": "pyannote/segmentation-3.0", "segmentation_model": "pyannote/segmentation-3.0",
"embedding_model": "pyannote/embedding", "embedding_model": "pyannote/embedding",

2
whisperlivekit/parse_args.py
View file

@ -61,7 +61,7 @@ def parse_args():
parser.add_argument( parser.add_argument(
"--diarization-backend", "--diarization-backend",
type=str, type=str,
default="diart", default="sortformer",
choices=["sortformer", "diart"], choices=["sortformer", "diart"],
help="The diarization backend to use.", help="The diarization backend to use.",
) )