Lettura di file audio con SpeechRecognition

Elaborazione del linguaggio parlato in Python

Daniel Bourke

Machine Learning Engineer/YouTube Creator

La classe AudioFile

import speech_recognition as sr

# Crea un'istanza del recognizer
recognizer = sr.Recognizer()

# Leggi il file audio
clean_support_call = sr.AudioFile("clean-support-call.wav")

# Controlla il tipo di clean_support_call
type(clean_support_call)

<class 'speech_recognition.AudioFile'>

Da AudioFile ad AudioData

recognizer.recognize_google(audio_data=clean_support_call)

AssertionError: ``audio_data`` must be audio data

# Converti da AudioFile a AudioData
with clean_support_call as source:

    # Registra l'audio
    clean_support_call_audio = recognizer.record(source)

# Controlla il tipo
type(clean_support_call_audio)

<class 'speech_recognition.AudioData'>

Trascrivere il nostro AudioData

# Trascrivi la chiamata pulita
recognizer.recognize_google(audio_data=clean_support_call_audio)

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

Duration e offset

duration e offset sono None per impostazione predefinita

# Lascia duration e offset ai valori predefiniti
with clean_support_call as source:
    clean_support_call_audio = recognizer.record(source,
                                                 duration=None,
                                                 offset=None)

# Primi 2 secondi della chiamata pulita
with clean_support_call as source:
    clean_support_call_audio = recognizer.record(source,
                                                 duration=2.0)

hello I'd like to get

Passons à la pratique !

Elaborazione del linguaggio parlato in Python