| | |
|---|
| | | <<<<<<< HEAD |
|---|
| | | import os |
|---|
| | | import shutil |
|---|
| | | import argparse |
|---|
| | |
|---|
| | | parser.add_argument('--gpuid', type=str, default="0") |
|---|
| | | args = parser.parse_args() |
|---|
| | | modelscope_infer(args) |
|---|
| | | ======= |
|---|
| | | from modelscope.pipelines import pipeline |
|---|
| | | from modelscope.utils.constant import Tasks |
|---|
| | | |
|---|
| | | if __name__ == "__main__": |
|---|
| | | audio_in = "https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_tr.wav" |
|---|
| | | output_dir = "./results" |
|---|
| | | inference_pipeline = pipeline( |
|---|
| | | task=Tasks.auto_speech_recognition, |
|---|
| | | model="damo/speech_UniASR_asr_2pass-tr-16k-common-vocab1582-pytorch", |
|---|
| | | output_dir=output_dir, |
|---|
| | | ) |
|---|
| | | rec_result = inference_pipeline(audio_in=audio_in, param_dict={"decoding_model":"offline"}) |
|---|
| | | print(rec_result) |
|---|
| | | >>>>>>> main |
|---|
