Skip to content

Commit a6bf059

Browse files
daspecsterdaspecster
authored
Rename and move _JSONSpeechAPI to HTTPSpeechAPI. (#2979)
* Rename and move _JSONSpeechAPI to HTTPSpeechAPI.
1 parent de65409 commit a6bf059

File tree

3 files changed

+214
-195
lines changed

3 files changed

+214
-195
lines changed

‎speech/google/cloud/speech/_http.py

Lines changed: 210 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,210 @@
1+
# Copyright 2017 Google Inc.
2+
#
3+
# Licensed under the Apache License, Version 2.0 (the "License");
4+
# you may not use this file except in compliance with the License.
5+
# You may obtain a copy of the License at
6+
#
7+
# http://www.apache.org/licenses/LICENSE-2.0
8+
#
9+
# Unless required by applicable law or agreed to in writing, software
10+
# distributed under the License is distributed on an "AS IS" BASIS,
11+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12+
# See the License for the specific language governing permissions and
13+
# limitations under the License.
14+
15+
"""HTTP module for managing Speech API requests."""
16+
17+
from base64 import b64encode
18+
19+
from google.cloud._helpers import _bytes_to_unicode
20+
from google.cloud._helpers import _to_bytes
21+
22+
from google.cloud.speech.result import Result
23+
from google.cloud.speech.operation import Operation
24+
25+
26+
class HTTPSpeechAPI(object):
27+
"""Speech API for interacting with the HTTP version of the API.
28+
29+
:type client: :class:`google.cloud.core.client.Client`
30+
:param client: Instance of a ``Client`` object.
31+
"""
32+
def __init__(self, client):
33+
self._client = client
34+
self._connection = client._connection
35+
36+
def async_recognize(self, sample, language_code=None,
37+
max_alternatives=None, profanity_filter=None,
38+
speech_context=None):
39+
"""Asychronous Recognize request to Google Speech API.
40+
41+
.. _async_recognize: https://cloud.google.com/speech/reference/\
42+
rest/v1beta1/speech/asyncrecognize
43+
44+
See `async_recognize`_.
45+
46+
:type sample: :class:`~google.cloud.speech.sample.Sample`
47+
:param sample: Instance of ``Sample`` containing audio information.
48+
49+
:type language_code: str
50+
:param language_code: (Optional) The language of the supplied audio as
51+
BCP-47 language tag. Example: ``'en-GB'``.
52+
If omitted, defaults to ``'en-US'``.
53+
54+
:type max_alternatives: int
55+
:param max_alternatives: (Optional) Maximum number of recognition
56+
hypotheses to be returned. The server may
57+
return fewer than maxAlternatives.
58+
Valid values are 0-30. A value of 0 or 1
59+
will return a maximum of 1. Defaults to 1
60+
61+
:type profanity_filter: bool
62+
:param profanity_filter: If True, the server will attempt to filter
63+
out profanities, replacing all but the
64+
initial character in each filtered word with
65+
asterisks, e.g. ``'f***'``. If False or
66+
omitted, profanities won't be filtered out.
67+
68+
:type speech_context: list
69+
:param speech_context: A list of strings (max 50) containing words and
70+
phrases "hints" so that the speech recognition
71+
is more likely to recognize them. This can be
72+
used to improve the accuracy for specific words
73+
and phrases. This can also be used to add new
74+
words to the vocabulary of the recognizer.
75+
76+
:rtype: :class:`~google.cloud.speech.operation.Operation`
77+
:returns: Operation for asynchronous request to Google Speech API.
78+
"""
79+
data = _build_request_data(sample, language_code, max_alternatives,
80+
profanity_filter, speech_context)
81+
api_response = self._connection.api_request(
82+
method='POST', path='speech:asyncrecognize', data=data)
83+
84+
operation = Operation.from_dict(api_response, self._client)
85+
operation.caller_metadata['request_type'] = 'AsyncRecognize'
86+
return operation
87+
88+
def sync_recognize(self, sample, language_code=None, max_alternatives=None,
89+
profanity_filter=None, speech_context=None):
90+
"""Synchronous Speech Recognition.
91+
92+
.. _sync_recognize: https://cloud.google.com/speech/reference/\
93+
rest/v1beta1/speech/syncrecognize
94+
95+
See `sync_recognize`_.
96+
97+
:type sample: :class:`~google.cloud.speech.sample.Sample`
98+
:param sample: Instance of ``Sample`` containing audio information.
99+
100+
:type language_code: str
101+
:param language_code: (Optional) The language of the supplied audio as
102+
BCP-47 language tag. Example: ``'en-GB'``.
103+
If omitted, defaults to ``'en-US'``.
104+
105+
:type max_alternatives: int
106+
:param max_alternatives: (Optional) Maximum number of recognition
107+
hypotheses to be returned. The server may
108+
return fewer than maxAlternatives.
109+
Valid values are 0-30. A value of 0 or 1
110+
will return a maximum of 1. Defaults to 1
111+
112+
:type profanity_filter: bool
113+
:param profanity_filter: If True, the server will attempt to filter
114+
out profanities, replacing all but the
115+
initial character in each filtered word with
116+
asterisks, e.g. ``'f***'``. If False or
117+
omitted, profanities won't be filtered out.
118+
119+
:type speech_context: list
120+
:param speech_context: A list of strings (max 50) containing words and
121+
phrases "hints" so that the speech recognition
122+
is more likely to recognize them. This can be
123+
used to improve the accuracy for specific words
124+
and phrases. This can also be used to add new
125+
words to the vocabulary of the recognizer.
126+
127+
:rtype: list
128+
:returns: A list of dictionaries. One dict for each alternative. Each
129+
dictionary typically contains two keys (though not
130+
all will be present in all cases)
131+
132+
* ``transcript``: The detected text from the audio recording.
133+
* ``confidence``: The confidence in language detection, float
134+
between 0 and 1.
135+
136+
:raises: ValueError if more than one result is returned or no results.
137+
"""
138+
data = _build_request_data(sample, language_code, max_alternatives,
139+
profanity_filter, speech_context)
140+
api_response = self._connection.api_request(
141+
method='POST', path='speech:syncrecognize', data=data)
142+
143+
if len(api_response['results']) > 0:
144+
results = api_response['results']
145+
return [Result.from_api_repr(result) for result in results]
146+
else:
147+
raise ValueError('No results were returned from the API')
148+
149+
150+
def _build_request_data(sample, language_code=None, max_alternatives=None,
151+
profanity_filter=None, speech_context=None):
152+
"""Builds the request data before making API request.
153+
154+
:type sample: :class:`~google.cloud.speech.sample.Sample`
155+
:param sample: Instance of ``Sample`` containing audio information.
156+
157+
:type language_code: str
158+
:param language_code: (Optional) The language of the supplied audio as
159+
BCP-47 language tag. Example: ``'en-GB'``.
160+
If omitted, defaults to ``'en-US'``.
161+
162+
:type max_alternatives: int
163+
:param max_alternatives: (Optional) Maximum number of recognition
164+
hypotheses to be returned. The server may
165+
return fewer than maxAlternatives.
166+
Valid values are 0-30. A value of 0 or 1
167+
will return a maximum of 1. Defaults to 1
168+
169+
:type profanity_filter: bool
170+
:param profanity_filter: If True, the server will attempt to filter
171+
out profanities, replacing all but the
172+
initial character in each filtered word with
173+
asterisks, e.g. ``'f***'``. If False or
174+
omitted, profanities won't be filtered out.
175+
176+
:type speech_context: list
177+
:param speech_context: A list of strings (max 50) containing words and
178+
phrases "hints" so that the speech recognition
179+
is more likely to recognize them. This can be
180+
used to improve the accuracy for specific words
181+
and phrases. This can also be used to add new
182+
words to the vocabulary of the recognizer.
183+
184+
:rtype: dict
185+
:returns: Dictionary with required data for Google Speech API.
186+
"""
187+
if sample.content is not None:
188+
audio = {'content':
189+
_bytes_to_unicode(b64encode(_to_bytes(sample.content)))}
190+
else:
191+
audio = {'uri': sample.source_uri}
192+
193+
config = {'encoding': sample.encoding,
194+
'sampleRate': sample.sample_rate}
195+
196+
if language_code is not None:
197+
config['languageCode'] = language_code
198+
if max_alternatives is not None:
199+
config['maxAlternatives'] = max_alternatives
200+
if profanity_filter is not None:
201+
config['profanityFilter'] = profanity_filter
202+
if speech_context is not None:
203+
config['speechContext'] = {'phrases': speech_context}
204+
205+
data = {
206+
'audio': audio,
207+
'config': config,
208+
}
209+
210+
return data

0 commit comments

Comments
 (0)
pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy