|
| 1 | +#!/usr/bin/env python |
| 2 | + |
| 3 | +# Copyright 2017 Google Inc. All Rights Reserved. |
| 4 | +# |
| 5 | +# Licensed under the Apache License, Version 2.0 (the "License"); |
| 6 | +# you may not use this file except in compliance with the License. |
| 7 | +# You may obtain a copy of the License at |
| 8 | +# |
| 9 | +# http://www.apache.org/licenses/LICENSE-2.0 |
| 10 | +# |
| 11 | +# Unless required by applicable law or agreed to in writing, software |
| 12 | +# distributed under the License is distributed on an "AS IS" BASIS, |
| 13 | +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 14 | +# See the License for the specific language governing permissions and |
| 15 | +# limitations under the License. |
| 16 | + |
| 17 | +"""Google Cloud Speech API sample that demonstrates how to select the model |
| 18 | +used for speech recognition. |
| 19 | +
|
| 20 | +Example usage: |
| 21 | + python transcribe_model_selection.py \ |
| 22 | + resources/Google_Gnome.wav --model video |
| 23 | + python transcribe_model_selection.py \ |
| 24 | + gs://cloud-samples-tests/speech/Google_Gnome.wav --model video |
| 25 | +""" |
| 26 | + |
| 27 | +import argparse |
| 28 | + |
| 29 | + |
| 30 | +# [START speech_transcribe_model_selection] |
| 31 | +def transcribe_model_selection(speech_file, model): |
| 32 | + """Transcribe the given audio file synchronously with |
| 33 | + the selected model.""" |
| 34 | + from google.cloud import speech_v1p1beta1 as speech |
| 35 | + client = speech.SpeechClient() |
| 36 | + |
| 37 | + with open(speech_file, 'rb') as audio_file: |
| 38 | + content = audio_file.read() |
| 39 | + |
| 40 | + audio = speech.types.RecognitionAudio(content=content) |
| 41 | + |
| 42 | + config = speech.types.RecognitionConfig( |
| 43 | + encoding=speech.enums.RecognitionConfig.AudioEncoding.LINEAR16, |
| 44 | + sample_rate_hertz=16000, |
| 45 | + language_code='en-US', |
| 46 | + model=model) |
| 47 | + |
| 48 | + response = client.recognize(config, audio) |
| 49 | + |
| 50 | + for i, result in enumerate(response.results): |
| 51 | + alternative = result.alternatives[0] |
| 52 | + print('-' * 20) |
| 53 | + print('First alternative of result {}'.format(i)) |
| 54 | + print('Transcript: {}'.format(alternative.transcript)) |
| 55 | +# [END speech_transcribe_model_selection] |
| 56 | + |
| 57 | + |
| 58 | +# [START speech_transcribe_model_selection_gcs] |
| 59 | +def transcribe_model_selection_gcs(gcs_uri, model): |
| 60 | + """Transcribe the given audio file asynchronously with |
| 61 | + the selected model.""" |
| 62 | + from google.cloud import speech_v1p1beta1 as speech |
| 63 | + client = speech.SpeechClient() |
| 64 | + |
| 65 | + audio = speech.types.RecognitionAudio(uri=gcs_uri) |
| 66 | + |
| 67 | + config = speech.types.RecognitionConfig( |
| 68 | + encoding=speech.enums.RecognitionConfig.AudioEncoding.LINEAR16, |
| 69 | + sample_rate_hertz=16000, |
| 70 | + language_code='en-US', |
| 71 | + model=model) |
| 72 | + |
| 73 | + operation = client.long_running_recognize(config, audio) |
| 74 | + |
| 75 | + print('Waiting for operation to complete...') |
| 76 | + response = operation.result(timeout=90) |
| 77 | + |
| 78 | + for i, result in enumerate(response.results): |
| 79 | + alternative = result.alternatives[0] |
| 80 | + print('-' * 20) |
| 81 | + print('First alternative of result {}'.format(i)) |
| 82 | + print('Transcript: {}'.format(alternative.transcript)) |
| 83 | +# [END speech_transcribe_model_selection_gcs] |
| 84 | + |
| 85 | + |
| 86 | +if __name__ == '__main__': |
| 87 | + parser = argparse.ArgumentParser( |
| 88 | + description=__doc__, |
| 89 | + formatter_class=argparse.RawDescriptionHelpFormatter) |
| 90 | + parser.add_argument( |
| 91 | + 'path', help='File or GCS path for audio file to be recognized') |
| 92 | + parser.add_argument( |
| 93 | + '--model', help='The speech recognition model to use', |
| 94 | + choices=['command_and_search', 'phone_call', 'video', 'default'], |
| 95 | + default='default') |
| 96 | + |
| 97 | + args = parser.parse_args() |
| 98 | + |
| 99 | + if args.path.startswith('gs://'): |
| 100 | + transcribe_model_selection_gcs(args.path, args.model) |
| 101 | + else: |
| 102 | + transcribe_model_selection(args.path, args.model) |
0 commit comments