In [9]:

import sys
sys.path.insert(0, '/home/ons21553/wspace/interview-transcripts/src')
from common import *

%load_ext autoreload
%autoreload 2

The autoreload extension is already loaded. To reload it, use:
  %reload_ext autoreload

In [1]:

import io
import os

# Imports the Google Cloud client library
from google.cloud import speech
from google.cloud.speech import enums
from google.cloud.speech import types

In [5]:

from common import *

In [27]:

os.environ['GOOGLE_APPLICATION_CREDENTIALS'] = from_root('dcm-iw-transcripts-db3f1fa338cc.json')

In [28]:

# Instantiates a client
client = speech.SpeechClient()

In [29]:

import recordings.harvard as hv
r = hv.load_all()[0]

In [30]:

with io.open(r.audio_fpath, 'rb') as audio_file:
    content = audio_file.read()
    audio = types.RecognitionAudio(content=content)

In [31]:

config = types.RecognitionConfig(
    encoding=enums.RecognitionConfig.AudioEncoding.LINEAR16,
    language_code='en-US')

# Detects speech in the audio file
response = client.recognize(config, audio)

In [34]:

r.structured_transcript

Out[34]:

	text	speaker	from	end
0	The small pup gnawed a hole in the sock.	NaN	NaN	NaN
1	The fish twisted and turned on the bent hook.	NaN	NaN	NaN
2	Press the pants and sew a button on the vest.	NaN	NaN	NaN
3	The swan dive was far short of perfect.	NaN	NaN	NaN
4	The beauty of the view stunned the young boy.	NaN	NaN	NaN
5	Two blue fish swam in the tank.	NaN	NaN	NaN
6	Her purse was full of useless trash.	NaN	NaN	NaN
7	The colt reared and threw the tall rider.	NaN	NaN	NaN
8	It snowed, rained, and hailed the same morning.	NaN	NaN	NaN
9	Read verse out loud for pleasure.	NaN	NaN	NaN

In [35]:

response

Out[35]:

results {
  alternatives {
    transcript: "The small pop not a hole in the sock the fish twisted and turned on the bent hook."
    confidence: 0.7337599396705627
  }
}
results {
  alternatives {
    transcript: " Pasta passes so it button on the vest."
    confidence: 0.6798253059387207
  }
}
results {
  alternatives {
    transcript: " The Swan Dive was Far short of perfect."
    confidence: 0.738798975944519
  }
}
results {
  alternatives {
    transcript: " the beauty of the views done the young boy"
    confidence: 0.742624819278717
  }
}
results {
  alternatives {
    transcript: " to blue fish swim in the tank"
    confidence: 0.7812932133674622
  }
}
results {
  alternatives {
    transcript: " her purse was full of useless trash."
    confidence: 0.7841848731040955
  }
}
results {
  alternatives {
    transcript: " The Colt reader does through the tall rider."
    confidence: 0.6258752942085266
  }
}
results {
  alternatives {
    transcript: " Is snow rain and hail the same morning?"
    confidence: 0.6540632247924805
  }
}
results {
  alternatives {
    transcript: " Read burst out loud for pleasure."
    confidence: 0.6588537693023682
  }
}

In [26]:

response.results[4]

Out[26]:

alternatives {
  transcript: " to blue fish swim in the tank"
  confidence: 0.7812932133674622
}

In [ ]:

for result in response.results:
    print('Transcript: {}'.format(result.alternatives[0].transcript))

In [ ]:

from google.cloud import speech
client = speech.SpeechClient()

with open(speech_file, 'rb') as audio_file:
    content = audio_file.read()

audio = speech.types.RecognitionAudio(content=content)

config = speech.types.RecognitionConfig(
    encoding=speech.enums.RecognitionConfig.AudioEncoding.LINEAR16,
    sample_rate_hertz=44100,
    language_code='en-US',
    audio_channel_count=2,
    enable_separate_recognition_per_channel=True)

response = client.recognize(config, audio)

for i, result in enumerate(response.results):
    alternative = result.alternatives[0]
    print('-' * 20)
    print('First alternative of result {}'.format(i))
    print(u'Transcript: {}'.format(alternative.transcript))
    print(u'Channel Tag: {}'.format(result.channel_tag))