SpeechRecognitionで音声ファイルを読み込む

Pythonで学ぶ音声言語処理

Daniel Bourke

Machine Learning Engineer/YouTube Creator

AudioFile クラス

import speech_recognition as sr

# 認識器を用意
recognizer = sr.Recognizer()

# 音声ファイルを読み込む
clean_support_call = sr.AudioFile("clean-support-call.wav")

# clean_support_call の型を確認
type(clean_support_call)

<class 'speech_recognition.AudioFile'>

AudioFile から AudioData へ

recognizer.recognize_google(audio_data=clean_support_call)

AssertionError: ``audio_data`` must be audio data

# AudioFile を AudioData に変換
with clean_support_call as source:

    # 音声を取得
    clean_support_call_audio = recognizer.record(source)

# 型を確認
type(clean_support_call_audio)

<class 'speech_recognition.AudioData'>

AudioData を文字起こし

# クリーンなサポート通話を文字起こし
recognizer.recognize_google(audio_data=clean_support_call_audio)

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

duration と offset

duration と offset は既定でどちらも None

# duration/offset は既定値のまま
with clean_support_call as source:
    clean_support_call_audio = recognizer.record(source,
                                                 duration=None,
                                                 offset=None)

# 先頭2秒のみを取得
with clean_support_call as source:
    clean_support_call_audio = recognizer.record(source,
                                                 duration=2.0)

hello I'd like to get

演習に進みましょう！

Pythonで学ぶ音声言語処理