befuturepl
/

embedding

pyannote-audio-model

speaker-recognition

speaker-verification

speaker-identification

speaker-embedding

Model card Files Files and versions

embedding / config.yaml

befuturepl's picture

Initial commit: diarization model

e69e0a4 over 1 year ago

History Blame Contribute Delete

221 Bytes

	model:
	_target_: pyannote.audio.models.embedding.WeSpeakerResNet34
	sample_rate: 16000
	num_channels: 1
	num_mel_bins: 80
	frame_length: 25
	frame_shift: 10
	dither: 0.0
	window_type: hamming
	use_energy: false