from modelscope.pipelines import pipeline
|
from modelscope.utils.constant import Tasks
|
|
|
if __name__ == '__main__':
|
param_dict = dict()
|
param_dict['hotword'] = "https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/hotword.txt"
|
|
audio_in = "//isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_hotword.wav"
|
output_dir = None
|
batch_size = 1
|
|
inference_pipeline = pipeline(
|
task=Tasks.auto_speech_recognition,
|
model="damo/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404",
|
output_dir=output_dir,
|
batch_size=batch_size,
|
param_dict=param_dict)
|
|
rec_result = inference_pipeline(audio_in=audio_in)
|
print(rec_result)
|
