提交

kongdeqiang

6 天以前 28ccfbfc51068a663a80764e14074df5edf2b5ba

 examples/industrial_data_pretraining/lcbnet/demo.py

@@ -5,30 +5,15 @@

from funasr import AutoModel

model = AutoModel(model="iic/LCB-NET")
model = AutoModel(model="iic/LCB-NET", model_revision="v1.0.0")


# example1
#res = model.generate(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav",
#                    )
#print(res)
res = model.generate(
    input=(
        "https://www.modelscope.cn/api/v1/models/iic/LCB-NET/repo?Revision=master&FilePath=example/asr_example.wav",
        "https://www.modelscope.cn/api/v1/models/iic/LCB-NET/repo?Revision=master&FilePath=example/ocr.txt",
    ),
    data_type=("sound", "text"),
)


'''
# tensor or numpy as input
# example2
import torchaudio
import os
wav_file = os.path.join(model.model_path, "example/asr_example.wav")
input_tensor, sample_rate = torchaudio.load(wav_file)
input_tensor = input_tensor.mean(0)
res = model.generate(input=[input_tensor], batch_size_s=300, is_final=True)


# example3
import soundfile

wav_file = os.path.join(model.model_path, "example/asr_example.wav")
speech, sample_rate = soundfile.read(wav_file)
res = model.generate(input=[speech], batch_size_s=300, is_final=True)
'''
print(res)