def deep_model(model: str = 'speakernet', quantized: bool = False, **kwargs):
"""
Load Speaker2Vec model.
Parameters
----------
model : str, optional (default='speakernet')
Model architecture supported. Allowed values:
* ``'vggvox-v1'`` - VGGVox V1, embedding size 1024, exported from https://github.com/linhdvu14/vggvox-speaker-identification
* ``'vggvox-v2'`` - VGGVox V2, embedding size 512, exported from https://github.com/WeidiXie/VGG-Speaker-Recognition
* ``'deep-speaker'`` - Deep Speaker, embedding size 512, exported from https://github.com/philipperemy/deep-speaker
* ``'speakernet'`` - SpeakerNet, embedding size 7205, exported from https://github.com/NVIDIA/NeMo/tree/main/examples/speaker_recognition
quantized : bool, optional (default=False)
if True, will load 8-bit quantized model.
Quantized model not necessary faster, totally depends on the machine.
Returns
-------
result : malaya_speech.supervised.classification.load function
"""
model = model.lower()
if model not in _availability:
raise Exception(
'model not supported, please check supported models from `malaya_speech.speaker_vector.available_model()`.'
)
return classification.load(model=model,
module='speaker-vector',
extra={},
label={},
quantized=quantized,
**kwargs)
def deep_model(model: str = 'vggvox-v2', quantized: bool = False, **kwargs):
"""
Load VAD model.
Parameters
----------
model : str, optional (default='vggvox-v2')
Model architecture supported. Allowed values:
* ``'vggvox-v1'`` - finetuned VGGVox V1.
* ``'vggvox-v2'`` - finetuned VGGVox V2.
quantized : bool, optional (default=False)
if True, will load 8-bit quantized model.
Quantized model not necessary faster, totally depends on the machine.
Returns
-------
result : malaya_speech.supervised.classification.load function
"""
model = model.lower()
if model not in _availability:
raise Exception(
'model not supported, please check supported models from `malaya_speech.vad.available_model()`.'
)
settings = {
'vggvox-v1': {
'frame_len': 0.005,
'frame_step': 0.0005
},
'vggvox-v2': {
'hop_length': 24,
'concat': False,
'mode': 'eval'
},
'speakernet': {
'frame_ms': 20,
'stride_ms': 1.0
},
}
return classification.load(model=model,
module='vad',
extra=settings[model],
label=[False, True],
quantized=quantized,
**kwargs)
def deep_model(model: str = 'speakernet', quantized: bool = False, **kwargs):
"""
Load speaker change deep model.
Parameters
----------
model : str, optional (default='vggvox-v2')
Model architecture supported. Allowed values:
* ``'vggvox-v2'`` - finetuned VGGVox V2.
* ``'speakernet'`` - finetuned SpeakerNet.
quantized : bool, optional (default=False)
if True, will load 8-bit quantized model.
Quantized model not necessary faster, totally depends on the machine.
Returns
-------
result : malaya_speech.supervised.classification.load function
"""
model = model.lower()
if model not in _availability:
raise Exception(
'model not supported, please check supported models from `malaya_speech.speaker_change.available_model()`.'
)
settings = {
'vggvox-v2': {
'hop_length': 50,
'concat': False,
'mode': 'eval'
},
'speakernet': {
'frame_ms': 20,
'stride_ms': 2
},
}
return classification.load(path=PATH_SPEAKER_CHANGE,
s3_path=S3_PATH_SPEAKER_CHANGE,
model=model,
name='speaker-change',
extra=settings[model],
label=[False, True],
quantized=quantized,
**kwargs)
def deep_model(model: str = 'vggvox-v2', quantized: bool = False, **kwargs):
"""
Load age detection deep model.
Parameters
----------
model : str, optional (default='vggvox-v2')
Model architecture supported. Allowed values:
* ``'vggvox-v2'`` - finetuned VGGVox V2.
* ``'deep-speaker'`` - finetuned Deep Speaker.
quantized : bool, optional (default=False)
if True, will load 8-bit quantized model.
Quantized model not necessary faster, totally depends on the machine.
Returns
-------
result : malaya_speech.supervised.classification.load function
"""
model = model.lower()
if model not in _availability:
raise Exception(
'model not supported, please check supported models from `malaya_speech.age_detection.available_model()`.'
)
settings = {
'vggvox-v2': {
'concat': False
},
'deep-speaker': {
'voice_only': False
},
}
return classification.load(path=PATH_AGE_DETECTION,
s3_path=S3_PATH_AGE_DETECTION,
model=model,
name='age-detection',
extra=settings[model],
label=labels,
quantized=quantized,
**kwargs)