114 lines
2.5 KiB
Python
114 lines
2.5 KiB
Python
# -*- coding: utf-8 -*-
|
|
"""
|
|
音频处理工具
|
|
"""
|
|
|
|
from pathlib import Path
|
|
from typing import Tuple
|
|
import numpy as np
|
|
import soundfile as sf
|
|
|
|
|
|
def load_audio_file(
|
|
audio_path: Path,
|
|
target_sample_rate: int = 16000,
|
|
) -> Tuple[bytes, float]:
|
|
"""
|
|
加载音频文件并转换为 PCM 16-bit 格式
|
|
|
|
Args:
|
|
audio_path: 音频文件路径
|
|
target_sample_rate: 目标采样率
|
|
|
|
Returns:
|
|
(pcm_bytes, duration_seconds): PCM 字节数据和音频时长(秒)
|
|
"""
|
|
# 读取音频文件
|
|
audio_data, sample_rate = sf.read(audio_path, dtype="float32")
|
|
|
|
# 如果是立体声，转换为单声道
|
|
if len(audio_data.shape) > 1:
|
|
audio_data = np.mean(audio_data, axis=1)
|
|
|
|
# 重采样到目标采样率
|
|
if sample_rate != target_sample_rate:
|
|
audio_data = resample_audio(audio_data, sample_rate, target_sample_rate)
|
|
|
|
# 计算时长
|
|
duration_seconds = len(audio_data) / target_sample_rate
|
|
|
|
# 转换为 16-bit PCM
|
|
pcm_data = (audio_data * 32767).astype(np.int16)
|
|
pcm_bytes = pcm_data.tobytes()
|
|
|
|
return pcm_bytes, duration_seconds
|
|
|
|
|
|
def resample_audio(
|
|
audio_data: np.ndarray,
|
|
orig_sample_rate: int,
|
|
target_sample_rate: int,
|
|
) -> np.ndarray:
|
|
"""
|
|
重采样音频
|
|
|
|
Args:
|
|
audio_data: 音频数据
|
|
orig_sample_rate: 原始采样率
|
|
target_sample_rate: 目标采样率
|
|
|
|
Returns:
|
|
重采样后的音频数据
|
|
"""
|
|
if orig_sample_rate == target_sample_rate:
|
|
return audio_data
|
|
|
|
# 计算重采样比例
|
|
ratio = target_sample_rate / orig_sample_rate
|
|
new_length = int(len(audio_data) * ratio)
|
|
|
|
# 使用线性插值进行重采样
|
|
x_old = np.linspace(0, 1, len(audio_data))
|
|
x_new = np.linspace(0, 1, new_length)
|
|
resampled = np.interp(x_new, x_old, audio_data)
|
|
|
|
return resampled.astype(np.float32)
|
|
|
|
|
|
def get_audio_duration(audio_path: Path) -> float:
|
|
"""
|
|
获取音频文件时长
|
|
|
|
Args:
|
|
audio_path: 音频文件路径
|
|
|
|
Returns:
|
|
时长(秒)
|
|
"""
|
|
info = sf.info(audio_path)
|
|
return info.duration
|
|
|
|
|
|
def split_audio_into_chunks(
|
|
pcm_bytes: bytes,
|
|
chunk_size: int,
|
|
) -> list:
|
|
"""
|
|
将 PCM 数据分割成块
|
|
|
|
Args:
|
|
pcm_bytes: PCM 字节数据
|
|
chunk_size: 每块的采样数
|
|
|
|
Returns:
|
|
字节块列表
|
|
"""
|
|
chunk_bytes = chunk_size * 2 # 16-bit = 2 bytes per sample
|
|
chunks = []
|
|
|
|
for i in range(0, len(pcm_bytes), chunk_bytes):
|
|
chunk = pcm_bytes[i : i + chunk_bytes]
|
|
chunks.append(chunk)
|
|
|
|
return chunks
|