2023-05-13 15:45:56 +00:00
|
|
|
class SpeechEncoder(object):
|
2023-05-14 06:39:07 +00:00
|
|
|
def __init__(self,vec_path = "pretrain/checkpoint_best_legacy_500.pt",device=None):
|
2023-05-13 15:45:56 +00:00
|
|
|
self.model = None #This is Model
|
|
|
|
self.hidden_dim = 768
|
|
|
|
pass
|
|
|
|
|
|
|
|
def encoder(self,wav):
|
|
|
|
'''
|
|
|
|
input: wav:[batchsize,signal_length]
|
|
|
|
output: embedding:[batchsize,wav_frame,hidden_dim]
|
|
|
|
'''
|
|
|
|
pass
|