处理音频数据单声道16k

why？

处理音频数据，模型需要处理为16k,单声道

So

import os
from pathlib import Path
from pydub import AudioSegment

# pydub处理
def convert_to_wav(src_path, dst_path, audio_format):
    sound = AudioSegment.from_file(src_path, format=audio_format).set_frame_rate(16000).set_channels(1)
    sound.export(dst_path, format="wav")

# ffmpeg处理
def cmd_convert_to_wav(src_path, dst_path):
    call = f'ffmpeg -y -i "{src_path}" -ar 16000 -ac 1 -f wav "{dst_path}"'
    os.system(call)


def batch_convert_to_wav(directory, filter_suffix=None, convert_suffix=None):
    filter_suffix = filter_suffix or []
    convert_suffix = convert_suffix or ['.m4a', '.mp3', '.wav']
    directory = Path(directory)
    for file in directory.iterdir():
        if not file.is_file():
            continue
        suffix = file.suffix.lower()
        if suffix in filter_suffix or suffix not in convert_suffix:
            continue
        wav_path = file.with_suffix('.wav')
        print(f"正在处理: {file}")
        try:
            convert_to_wav(str(file), str(wav_path), audio_format=suffix.lstrip('.'))
        except Exception as e:
            print(f"pydub转换失败: {e}，尝试ffmpeg...")
            cmd_convert_to_wav(str(file), str(wav_path))


if __name__ == "__main__":
    # 音频转为wav,单声道,16k
    _dir = r""
    batch_convert_to_wav(_dir, filter_suffix=[], convert_suffix=['.m4a', '.mp3', '.wav'])