function transcribe_file_with_word_time_offsets()

Fig.1 Flowchart of function transcribe_file_with_word_time_offsets

Fig.1呈現函數transcribe_file_with_word_time_offsets的流程圖。

函數transcribe_file_with_word_time_offsets的輸入參數包含$\colon$

• speech_file 音頻的檔名
• timstamp_writer xlsx的檔名
• language_code 轉譯的語言
• beta google cloud api beta版？
• alternative_language 替代語言(沒有使用)

函數transcribe_file_with_word_time_offsets有三大程式區塊

1. Setup Google Cloud Speech-to-Text types.RecognitionConfig
2. Run Google Cloud Speech Recognition
3. Parse Recognition Response and Save Result to excel file

函數transcribe_file_with_word_time_offsets把文件轉譯結果存進results變數裡輸出，在transcribe_file_with_word_time_offset.py裡，是返回函數batch_transcribe_speechFile (行143)。

045 def transcribe_file_with_word_time_offsets(speech_file, timestamp_writer, language_code='cmn-Hant-TW', beta=False, alternative_language= None):
046     """Transcribe the given audio file synchronously and output the word time
047     offsets."""
048     samp_freq, _ = wavfile.read(speech_file)
049     if beta:
050         from google.cloud import speech_v1p1beta1 as speech
051         from google.cloud.speech_v1p1beta1 import enums
052         from google.cloud.speech_v1p1beta1 import types
053 
054         config = types.RecognitionConfig(
055             encoding=enums.RecognitionConfig.AudioEncoding.LINEAR16,
056             sample_rate_hertz=samp_freq,
057             language_code=language_code,
058             #alternative_language_codes=['en-US'],
059             alternative_language_codes = alternative_language,
060             enable_word_time_offsets=True,
061             speech_contexts=[types.SpeechContext(
062                 phrases=['四', '三', '二', '一'],
063                 )],
064             enable_word_confidence=True
065             )
066 
067     else:        
068         from google.cloud import speech
069         from google.cloud.speech import enums
070         from google.cloud.speech import types
071 
072         config = types.RecognitionConfig(
073             encoding=enums.RecognitionConfig.AudioEncoding.LINEAR16,
074             sample_rate_hertz=samp_freq,
075             language_code=language_code,
076             enable_word_time_offsets=True,
077             speech_contexts=[types.SpeechContext(
078                 phrases=['四', '三', '二', '一'],
079                 )],
080             #enable_word_confidence=True
081             )
082     
083     client = speech.SpeechClient()
084     
085     with io.open(speech_file, 'rb') as audio_file:
086         content = audio_file.read()
087 
088     audio = types.RecognitionAudio(content=content)
089     
090     response = client.recognize(config, audio)
091     
092     _, filename = parse_filepath(speech_file)
093     n = re.match(r'C(.+)\.wav', filename)
094     audio_index = n.group(1)
095     results = []
096     for i, result in enumerate(response.results):
097         alternative = result.alternatives[0]
098         print(u'Transcript {}_{}: {}'.format(audio_index, i+1, alternative.transcript))
099         results.append(alternative.transcript)
100 
101         word_index = []
102         start_time_column = []
103         end_time_column   = []
104         confidences_list  = []
105         for word_info in alternative.words:
106             word = word_info.word
107             start_time = word_info.start_time
108             end_time = word_info.end_time
109             word_index.append(word)
110             start_time_column.append(start_time.seconds + start_time.nanos * 1e-9)
111             end_time_column.append(end_time.seconds + end_time.nanos * 1e-9)
112             if hasattr(word_info, 'confidence'):
113                 confidences_list.append(word_info.confidence)
114             else:
115                 confidences_list.append(float('nan'))
116                 
117         df = pd.DataFrame({'start_time':start_time_column, 'end_time':end_time_column, 'confidence':confidences_list}, index=word_index)
118         #print(df)
119         df.index.name = 'word'
120         df.to_excel(timestamp_writer,'{}_{}'.format(filename, i+1))
121     
122     return results