Automatic Speech Recognition
pyannote.audio
pyannote
pyannote-audio-pipeline
audio
voice
speech
speaker
speaker-diarization
speaker-change-detection
voice-activity-detection
overlapped-speech-detection
Instructions to use anilbs/pipeline with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- pyannote.audio
How to use anilbs/pipeline with pyannote.audio:
from pyannote.audio import Pipeline pipeline = Pipeline.from_pretrained("anilbs/pipeline") # inference on the whole file pipeline("file.wav") # inference on an excerpt from pyannote.core import Segment excerpt = Segment(start=2.0, end=5.0) from pyannote.audio import Audio waveform, sample_rate = Audio().crop("file.wav", excerpt) pipeline({"waveform": waveform, "sample_rate": sample_rate}) - Notebooks
- Google Colab
- Kaggle
| pipeline: | |
| name: pyannote.audio.pipelines.SpeakerDiarization | |
| params: | |
| segmentation: anilbs/segmentation | |
| segmentation_step: 0.1 | |
| segmentation_batch_size: 32 | |
| # embedding: speechbrain/spkrec-ecapa-voxceleb@5c0be3875fda05e81f3c004ed8c7c06be308de1e | |
| embedding: speechbrain/spkrec-ecapa-voxceleb | |
| embedding_exclude_overlap: True | |
| embedding_batch_size: 32 | |
| clustering: HiddenMarkovModelClustering | |
| params: | |
| segmentation_onset: 0.58 | |
| clustering: | |
| covariance_type: diag | |
| threshold: 0.35 | |
| single_cluster_detection: | |
| quantile: 0.05 | |
| threshold: 1.15 | |