|
|
@@ -5,10 +5,10 @@ from modelscope.utils.constant import Tasks
|
|
|
|
|
|
input = 'https://modelscope.cn/api/v1/models/damo/speech_separation_mossformer_8k_pytorch/repo?Revision=master&FilePath=examples/mix_speech1.wav'
|
|
|
separation = pipeline(
|
|
|
- Tasks.funasr_speech_separation,
|
|
|
+ Tasks.speech_separation,
|
|
|
model='damo/speech_separation_mossformer_8k_pytorch',
|
|
|
output_dir='./',
|
|
|
- model_revision='v1.0.1')
|
|
|
+ model_revision='v1.0.2')
|
|
|
result = separation(audio_in=input)
|
|
|
for i, signal in enumerate(result):
|
|
|
save_file = f'output_spk_{i+1}.wav'
|
hnluo