参考

不OK的项目

speech_recognition

基础

视频提取音频

# -*- coding:utf-8 _*-
"""
Author: 
Email: 
Date: 2021/03/23
File: audio_test.py
Software: PyCharm
Description: 音频识别
"""


# load modules
import os
import sys
from moviepy.editor import AudioFileClip

current_path = sys.path[0]
video_path = '%s/video/' % current_path  # 视频源文件目录
video_name = os.listdir(video_path)[0]
audio_path = '%s/audio_extracts/' % current_path  # 音频提取文件目录
if not os.path.exists(audio_path):
    os.mkdir(audio_path)
audio_name = video_name.split('.')[0] + '.wav'

# 音频提取
my_audio_clip = AudioFileClip(video_path+video_name)
my_audio_clip.write_audiofile(audio_path+audio_name)

# 音频分析
import numpy as np
import librosa
audio, freq = librosa.load(audio_path+audio_name)
time = np.arange(0, len(audio)) / freq
print(len(audio), type(audio), freq, sep="\t")

# 画信号增强图
import librosa.display
import matplotlib.pyplot as plt
audio, _ = librosa.effects.trim(audio)  # Trim leading and trailing #silence from an audio signal.
librosa.display.waveplot(audio, sr=freq)
plt.show()