tools/cmd-denoise.py

import os,argparse

from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks
from tqdm import tqdm

path_denoise  = 'tools/denoise-model/speech_frcrn_ans_cirm_16k'
path_denoise  = path_denoise  if os.path.exists(path_denoise)  else "damo/speech_frcrn_ans_cirm_16k"
ans = pipeline(Tasks.acoustic_noise_suppression,model=path_denoise)
def execute_denoise(input_folder,output_folder):
    os.makedirs(output_folder,exist_ok=True)
    # print(input_folder)
    # print(list(os.listdir(input_folder).sort()))
    for name in tqdm(os.listdir(input_folder)):
        ans("%s/%s"%(input_folder,name),output_path='%s/%s'%(output_folder,name))

if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument("-i", "--input_folder", type=str, required=True,
                        help="Path to the folder containing WAV files.")
    parser.add_argument("-o", "--output_folder", type=str, required=True, 
                        help="Output folder to store transcriptions.")
    parser.add_argument("-p", "--precision", type=str, default='float16', choices=['float16','float32'],
                        help="fp16 or fp32")#还没接入
    cmd = parser.parse_args()
    execute_denoise(
        input_folder  = cmd.input_folder,
        output_folder = cmd.output_folder,
    )