Files

2.0 KiB

In [ ]:
from pyannote.database import get_protocol, FileFinder
protocol = get_protocol('Debug.SpeakerDiarization.Debug', 
                        preprocessors={"audio": FileFinder()})
In [ ]:
from pyannote.audio.tasks import VoiceActivityDetection
from pyannote.audio.models.segmentation.debug import SimpleSegmentationModel
import pytorch_lightning as pl
In [ ]:
vad = VoiceActivityDetection(protocol, duration=2., batch_size=16, num_workers=4)
model = SimpleSegmentationModel(task=vad)
trainer = pl.Trainer(max_epochs=1)
_ = trainer.fit(model)
In [ ]:
summary = model.summarize('full')
In [ ]:
model.freeze_up_to('lstm')
In [ ]:
model.unfreeze_up_to('mfcc.MelSpectrogram.spectrogram')
In [ ]:
model.freeze_by_name(['lstm', 'activation'])