Omgaan met verschillende soorten audio

Spoken Language Processing in Python

Daniel Bourke

Machine Learning Engineer/YouTube Creator

Welke taal?

# Maak een recognizer-klasse
recognizer = sr.Recognizer()

# Geef de Japanse audio door aan recognize_google
text = recognizer.recognize_google(japanese_good_morning, 
                                   language="en-US")

# Print de tekst
print(text)

Ohio gozaimasu

Welke taal?

# Maak een recognizer-klasse
recognizer = sr.Recognizer()

# Geef de Japanse audio door aan recognize_google
text = recognizer.recognize_google(japanese_good_morning, 
                                   language="ja")

# Print de tekst
print(text)

おはようございます

Niet-spraakaudio

# Importeer het audiofragment van een luipaardbrul
leopard_roar = sr.AudioFile("leopard_roar.wav")

# Zet de AudioFile om naar AudioData
with leopard_roar as source:
    leopard_roar_audio = recognizer.record(source)

# Herken de AudioData
recognizer.recognize_google(leopard_roar_audio)

UnknownValueError:

Niet-spraakaudio

# Importeer het audiofragment van een luipaardbrul
leopard_roar = sr.AudioFile("leopard_roar.wav")

# Zet de AudioFile om naar AudioData
with leopard_roar as source:
    leopard_roar_audio = recognizer.record(source)

# Herken de AudioData met show_all ingeschakeld
recognizer.recognize_google(leopard_roar_audio,
                            show_all=True)

[]

Alles tonen

# Japanse audio herkennen met show_all=True
text = recognizer.recognize_google(japanese_good_morning, 
                                   language="en-US",
                                   show_all=True)
# Print de tekst
print(text)

{'alternative': [{'transcript': 'Ohio gozaimasu', 'confidence': 0.89041114},
  {'transcript': 'all hail gozaimasu'},
  {'transcript': 'ohayo gozaimasu'},
  {'transcript': 'olho gozaimasu'},
  {'transcript': 'all Hale gozaimasu'}],
 'final': True}

Meerdere sprekers

# Importeer een audiobestand met meerdere sprekers
multiple_speakers = sr.AudioFile("multiple-speakers.wav")

# Zet AudioFile om naar AudioData
with multiple_speakers as source:
    multiple_speakers_audio = recognizer.record(source)

# Herken de AudioData
recognizer.recognize_google(multiple_speakers_audio)

one of the limitations of the speech recognition library is that it doesn't
recognise different speakers and voices it will just return it all as one block 
of text

Meerdere sprekers

# Importeer audiobestanden apart
speakers = [sr.AudioFile("s0.wav"), sr.AudioFile("s1.wav"), sr.AudioFile("s2.wav")]

# Transcribeer elke spreker afzonderlijk
for i, speaker in enumerate(speakers):
    with speaker as source:
        speaker_audio = recognizer.record(source)
    print(f"Text from speaker {i}: {recognizer.recognize_google(speaker_audio)}")

Text from speaker 0: one of the limitations of the speech recognition library
Text from speaker 1: is that it doesn't recognise different speakers and voices
Text from speaker 2: it will just return it all as one block a text

Ruis in audio

Als jij moeite hebt om het te verstaan, dan de API's ook

# Importeer een audiobestand met achtergrondgeluid
noisy_support_call = sr.AudioFile("noisy_support_call.wav")

with noisy_support_call as source:
    # Corrigeer voor omgevingsgeluid en neem op
    recognizer.adjust_for_ambient_noise(source,
                                        duration=0.5)
    noisy_support_call_audio = recognizer.record(source)

# Herken de audio
recognizer.recognize_google(noisy_support_call_audio)

hello ID like to get some help setting up my calories

Laten we oefenen!

Spoken Language Processing in Python