egs_modelscope/tp/speech_timestamp_prediction-v1-16k-offline/demo_long.py
@@ -8,9 +8,9 @@ import soundfile as sf param_dict = dict() param_dict['hotword'] = "信诺" param_dict['hotword'] = "你的热词" test_wav = '/Users/shixian/Downloads/tpdebug.wav' test_wav = 'YOUR_LONG_WAV.wav' output_dir = './tmp' os.system("mkdir -p {}".format(output_dir))
