Files
pyannote-audio/notebook/example.ipynb
T

4.8 KiB

In [ ]:
from pyannote.database import get_protocol, FileFinder
protocol = get_protocol('Debug.SpeakerDiarization.Debug', 
                        preprocessors={"audio": FileFinder()})
In [ ]:
from pyannote.audio.models.segmentation.debug import SimpleSegmentationModel
from pyannote.audio.models.segmentation.debug import MultiTaskSegmentationModel
from pyannote.audio.models.embedding.debug import SimpleEmbeddingModel
import pytorch_lightning as pl
In [ ]:
from pyannote.audio.utils.preview import listen
listen(next(protocol.train()))

## Voice activity detection

In [ ]:
from pyannote.audio.tasks import VoiceActivityDetection
vad = VoiceActivityDetection(protocol, duration=2., batch_size=32, num_workers=4)
model = SimpleSegmentationModel(task=vad)
trainer = pl.Trainer(max_epochs=1)
_ = trainer.fit(model)

## Speaker change detection

In [ ]:
from pyannote.audio.tasks import SpeakerChangeDetection
scd = SpeakerChangeDetection(protocol, duration=2., batch_size=32, num_workers=4)
model = SimpleSegmentationModel(task=scd)
trainer = pl.Trainer(max_epochs=1)
_ = trainer.fit(model)

Overlapped speech detection

In [ ]:
from pyannote.audio.tasks import OverlappedSpeechDetection
ovl = OverlappedSpeechDetection(protocol, duration=2., batch_size=32, num_workers=4)
model = SimpleSegmentationModel(task=ovl)
trainer = pl.Trainer(max_epochs=1)
_ = trainer.fit(model)

Segmentation

In [ ]:
from pyannote.audio.tasks import Segmentation
seg = Segmentation(protocol, duration=2., batch_size=32, num_workers=4)
model = SimpleSegmentationModel(task=seg)
trainer = pl.Trainer(max_epochs=1)
_ = trainer.fit(model)

Multi-task segmentation

In [ ]:
from pyannote.audio.tasks import MultiTaskSegmentation
seg = MultiTaskSegmentation(protocol, vad=True, scd=True, osd=True, duration=2., batch_size=32, num_workers=4)
model = MultiTaskSegmentationModel(task=seg)
trainer = pl.Trainer(max_epochs=1)
_ = trainer.fit(model)

## Speaker tracking

In [ ]:
from pyannote.audio.tasks import SpeakerTracking
spk = SpeakerTracking(protocol, duration=2., batch_size=32, num_workers=4)
model = SimpleSegmentationModel(task=spk)
trainer = pl.Trainer(max_epochs=1)
_ = trainer.fit(model)

Speaker embedding

In [ ]:
from pyannote.audio.tasks import SpeakerEmbedding
emb = SpeakerEmbedding(protocol, duration=2., num_workers=4)
model = SimpleEmbeddingModel(task=emb)
trainer = pl.Trainer(max_epochs=1)
_ = trainer.fit(model)