處理不同類型的音訊

Python 的口語語言處理

Daniel Bourke

Machine Learning Engineer/YouTube Creator

是什麼語言？

# 建立辨識器類別
recognizer = sr.Recognizer()

# 將日文音訊傳給 recognize_google
text = recognizer.recognize_google(japanese_good_morning, 
                                   language="en-US")

# 印出文字
print(text)

Ohio gozaimasu

是什麼語言？

# 建立辨識器類別
recognizer = sr.Recognizer()

# 將日文音訊傳給 recognize_google
text = recognizer.recognize_google(japanese_good_morning, 
                                   language="ja")

# 印出文字
print(text)

おはようございます

非語音音訊

# 匯入豹吼音訊檔
leopard_roar = sr.AudioFile("leopard_roar.wav")

# 將 AudioFile 轉為 AudioData
with leopard_roar as source:
    leopard_roar_audio = recognizer.record(source)

# 語音辨識 AudioData
recognizer.recognize_google(leopard_roar_audio)

UnknownValueError:

非語音音訊

# 匯入豹吼音訊檔
leopard_roar = sr.AudioFile("leopard_roar.wav")

# 將 AudioFile 轉為 AudioData
with leopard_roar as source:
    leopard_roar_audio = recognizer.record(source)

# 開啟 show_all 來進行語音辨識
recognizer.recognize_google(leopard_roar_audio,
                            show_all=True)

[]

顯示全部

# 使用 show_all=True 辨識日文音訊
text = recognizer.recognize_google(japanese_good_morning, 
                                   language="en-US",
                                   show_all=True)
# 印出文字
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}

多位說話者

# 匯入含多位說話者的音訊檔
multiple_speakers = sr.AudioFile("multiple-speakers.wav")

# 將 AudioFile 轉為 AudioData
with multiple_speakers as source:
    multiple_speakers_audio = recognizer.record(source)

# 語音辨識 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

多位說話者

# 分別匯入各說話者的音訊檔
speakers = [sr.AudioFile("s0.wav"), sr.AudioFile("s1.wav"), sr.AudioFile("s2.wav")]

# 各自轉錄每位說話者
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

含雜訊音訊

如果連你都聽不清楚，API 也會聽不清楚。

# 匯入含背景雜訊的音訊檔
noisy_support_call = sr.AudioFile(noisy_support_call.wav)

with noisy_support_call as source:
    # 先校正環境雜訊再錄製
    recognizer.adjust_for_ambient_noise(source, 
                                        duration=0.5)
    noisy_support_call_audio = recognizer.record(source)

# 語音辨識音訊
recognizer.recognize_google(noisy_support_call_audio)

hello ID like to get some help setting up my calories

一起來練習吧！

Python 的口語語言處理