GeezerVison is showing github.com — python-docs-samples/speech/api/speech_streaming.py at master · wuser08/python-docs-samples

Expand file tree

speech_streaming.py

Copy path

More file actions

Latest commit

History

157 lines (129 loc) · 5.6 KB

speech_streaming.py

Copy path

File metadata and controls

157 lines (129 loc) · 5.6 KB

Raw

Copy raw file

Download raw file

Open symbols panel

Edit and raw actions

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

#!/usr/bin/python

# Licensed under the Apache License, Version 2.0 (the "License");

# you may not use this file except in compliance with the License.

# You may obtain a copy of the License at

# http://www.apache.org/licenses/LICENSE-2.0

# Unless required by applicable law or agreed to in writing, software

# distributed under the License is distributed on an "AS IS" BASIS,

# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.

# See the License for the specific language governing permissions and

# limitations under the License.

"""Sample that streams audio to the Google Cloud Speech API via GRPC."""

from __future__ import division

import contextlib

import re

import threading

from gcloud.credentials import get_credentials

from google.cloud.speech.v1beta1 import cloud_speech_pb2 as cloud_speech

from google.rpc import code_pb2

from grpc.beta import implementations

import pyaudio

# Audio recording parameters

RATE = 16000

CHANNELS = 1

CHUNK = int(RATE / 10) # 100ms

# Keep the request alive for this many seconds

DEADLINE_SECS = 8 * 60 * 60

SPEECH_SCOPE = 'https://www.googleapis.com/auth/cloud-platform'

def make_channel(host, port):

"""Creates an SSL channel with auth credentials from the environment."""

# In order to make an https call, use an ssl channel with defaults

ssl_channel = implementations.ssl_channel_credentials(None, None, None)

# Grab application default credentials from the environment

creds = get_credentials().create_scoped([SPEECH_SCOPE])

# Add a plugin to inject the creds into the header

auth_header = (

'Authorization',

'Bearer ' + creds.get_access_token().access_token)

auth_plugin = implementations.metadata_call_credentials(

lambda _, cb: cb([auth_header], None),

name='google_creds')

# compose the two together for both ssl and google auth

composite_channel = implementations.composite_channel_credentials(

ssl_channel, auth_plugin)

return implementations.secure_channel(host, port, composite_channel)

# [START audio_stream]

@contextlib.contextmanager

def record_audio(channels, rate, chunk):

"""Opens a recording stream in a context manager."""

audio_interface = pyaudio.PyAudio()

audio_stream = audio_interface.open(

format=pyaudio.paInt16, channels=channels, rate=rate,

input=True, frames_per_buffer=chunk,

)

yield audio_stream

audio_stream.stop_stream()

audio_stream.close()

audio_interface.terminate()

# [END audio_stream]

def request_stream(stop_audio, channels=CHANNELS, rate=RATE, chunk=CHUNK):

"""Yields `StreamingRecognizeRequest`s constructed from a recording audio

stream.

Args:

stop_audio: A threading.Event object stops the recording when set.

channels: How many audio channels to record.

rate: The sampling rate in hertz.

chunk: Buffer audio into chunks of this size before sending to the api.

"""

# The initial request must contain metadata about the stream, so the

# server knows how to interpret it.

recognition_config = cloud_speech.RecognitionConfig(

# There are a bunch of config options you can specify. See

# https://goo.gl/A6xv5G for the full list.

encoding='LINEAR16', # raw 16-bit signed LE samples

sample_rate=rate, # the rate in hertz

# See

# https://g.co/cloud/speech/docs/best-practices#language_support

# for a list of supported languages.

language_code='en-US', # a BCP-47 language tag

)

streaming_config = cloud_speech.StreamingRecognitionConfig(

config=recognition_config,

# Note that setting interim_results to True means that you'll likely

# get multiple results for the same bit of audio, as the system

# re-interprets audio in the context of subsequent audio. However, this

# will give us quick results without having to tell the server when to

# finalize a piece of audio.

interim_results=True, single_utterance=True

)

yield cloud_speech.StreamingRecognizeRequest(

streaming_config=streaming_config)

with record_audio(channels, rate, chunk) as audio_stream:

while not stop_audio.is_set():

data = audio_stream.read(chunk)

if not data:

raise StopIteration()

# Subsequent requests can all just have the content

yield cloud_speech.StreamingRecognizeRequest(audio_content=data)

def listen_print_loop(recognize_stream):

for resp in recognize_stream:

if resp.error.code != code_pb2.OK:

raise RuntimeError('Server error: ' + resp.error.message)

# Display the transcriptions & their alternatives

for result in resp.results:

print(result.alternatives)

# Exit recognition if any of the transcribed phrases could be

# one of our keywords.

if any(re.search(r'\b(exit|quit)\b', alt.transcript)

for result in resp.results

for alt in result.alternatives):

print('Exiting..')

return

def main():

stop_audio = threading.Event()

with cloud_speech.beta_create_Speech_stub(

make_channel('speech.googleapis.com', 443)) as service:

try:

listen_print_loop(

service.StreamingRecognize(

request_stream(stop_audio), DEADLINE_SECS))

finally:

# Stop the request stream once we're done with the loop - otherwise

# it'll keep going in the thread that the grpc lib makes for it..

stop_audio.set()

if __name__ == '__main__':

main()