使用 SpeechRecognition 读取音频文件

Python 语音语言处理

Daniel Bourke

Machine Learning Engineer/YouTube Creator

AudioFile 类

import speech_recognition as sr

# 设置识别器实例
recognizer = sr.Recognizer()

# 读取音频文件
clean_support_call = sr.AudioFile("clean-support-call.wav")

# 查看 clean_support_call 的类型
type(clean_support_call)

<class 'speech_recognition.AudioFile'>

从 AudioFile 到 AudioData

recognizer.recognize_google(audio_data=clean_support_call)

AssertionError: ``audio_data`` must be audio data

# 将 AudioFile 转为 AudioData
with clean_support_call as source:

    # 录制音频
    clean_support_call_audio = recognizer.record(source)

# 查看类型
type(clean_support_call_audio)

<class 'speech_recognition.AudioData'>

转写我们的 AudioData

# 转写干净的客服通话
recognizer.recognize_google(audio_data=clean_support_call_audio)

hello I'd like to get some help setting up my account please

duration 与 offset

duration 与 offset 默认均为 None

# 使用默认的 duration 和 offset
with clean_support_call as source:
    clean_support_call_audio = recognizer.record(source,
                                                 duration=None,
                                                 offset=None)

# 获取前 2 秒的通话
with clean_support_call as source:
    clean_support_call_audio = recognizer.record(source,
                                                 duration=2.0)

hello I'd like to get

让我们练习！

Python 语音语言处理