# Copyright 2016 IBM All Rights Reserved.
#
# 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.
"""
The v1 Speech to Text service
(https://www.ibm.com/watson/developercloud/speech-to-text.html)
"""
from .watson_service import WatsonService
import json
[docs]class SpeechToTextV1(WatsonService):
default_url = "https://stream.watsonplatform.net/speech-to-text/api"
def __init__(self, url=default_url, **kwargs):
WatsonService.__init__(self, 'speech_to_text', url, **kwargs)
[docs] def recognize(self, audio, content_type, continuous=None, model=None,
customization_id=None,
inactivity_timeout=None,
keywords=None, keywords_threshold=None,
max_alternatives=None,
word_alternatives_threshold=None,
word_confidence=None, timestamps=None, interim_results=None,
profanity_filter=None,
smart_formatting=None,
speaker_labels=None,
customization_weight=None):
"""
Returns the recognized text from the audio input
"""
headers = {'content-type': content_type}
params = {'continuous': continuous,
'inactivity_timeout': inactivity_timeout,
'keywords': keywords,
'keywords_threshold': keywords_threshold,
'max_alternatives': max_alternatives,
'model': model,
'customization_id': customization_id,
'word_alternatives_threshold': word_alternatives_threshold,
'word_confidence': word_confidence,
'timestamps': timestamps,
'interim_results': interim_results,
'profanity_filter': profanity_filter,
'smart_formatting': smart_formatting,
'speaker_labels': speaker_labels,
'customization_weight': customization_weight}
return self.request(method='POST', url='/v1/recognize',
headers=headers,
data=audio, params=params,
stream=True, accept_json=True)
[docs] def models(self):
"""
Returns the list of available models to use with recognize
"""
return self.request(method='GET', url='/v1/models', accept_json=True)
[docs] def get_model(self, model_id):
"""
:param model_id: The identifier of the desired model
:return: A single instance of a Model object with results for the
specified model.
"""
return self.request(method='GET',
url='/v1/models/{0}'.format(model_id),
accept_json=True)
[docs] def create_custom_model(self, name, description="",
base_model="en-US_BroadbandModel"):
json_body = json.dumps({'name': name, 'description': description,
'base_model_name': base_model})
return self.request(method='POST', url='/v1/customizations',
headers={'content-type': 'application/json'},
data=json_body, accept_json=True)
[docs] def train_custom_model(self, customization_id,
customization_weight=None,
word_type=None):
"""
Trains a custom language model
"""
params = {'customization_weight': customization_weight,
'word_type': word_type}
return self.request(method='POST',
url=('/v1/customizations/{0}/train'
.format(customization_id)), params=params,
accept_json=True)
[docs] def list_custom_models(self):
return self.request(method='GET', url='/v1/customizations',
accept_json=True)
[docs] def get_custom_model(self, modelid):
return self.request(method='GET',
url='/v1/customizations/{0}'.format(modelid),
accept_json=True)
[docs] def delete_custom_model(self, modelid):
return self.request(method='DELETE',
url='/v1/customizations/{0}'.format(modelid),
accept_json=True)
[docs] def list_corpora(self, customization_id):
url = '/v1/customizations/{0}/corpora'
return self.request(method='GET',
url=url.format(customization_id),
accept_json=True)
[docs] def add_corpus(self,
customization_id,
corpus_name,
file_data,
allow_overwrite=None):
url = '/v1/customizations/{0}/corpora/{1}'
if allow_overwrite is None:
allow_overwrite = False
headers = {'Content-Type': 'application/octet-stream'}
return self.request(method='POST',
url=url.format(customization_id,
corpus_name),
headers=headers,
data=file_data,
params={'allow_overwrite': allow_overwrite},
accept_json=True)
[docs] def get_corpus(self, customization_id, corpus_name):
url = '/v1/customizations/{0}/corpora/{1}'
return self.request(method='GET',
url=url.format(customization_id,
corpus_name),
accept_json=True)
[docs] def delete_corpus(self, customization_id, corpus_name):
url = '/v1/customizations/{0}/corpora/{1}'
return self.request(method='DELETE',
url=url.format(customization_id,
corpus_name),
accept_json=True)
[docs] class CustomWord(object):
def __init__(self, word=None, sounds_like=None, display_as=None):
self._word = word
self._sounds_like = sounds_like
self._display_as = display_as
@property
def word(self):
return self._word
@property
def sounds_like(self):
return self._sounds_like
@property
def display_as(self):
return self._display_as
def __dict__(self):
return {'word': self.word,
'sounds_like': self.sounds_like,
'display_as': self.display_as}
[docs] def add_custom_words(self, customization_id, custom_words):
url = '/v1/customizations/{0}/words'
payload = {'words': [x.__dict__() for x in custom_words]}
return self.request(method='POST',
url=url.format(customization_id),
data=json.dumps(payload),
headers={'content-type': 'application/json'},
accept_json=True)
[docs] def add_custom_word(self, customization_id, custom_word):
url = '/v1/customizations/{0}/words/{1}'
custom_word_fragment = {'sounds_like': custom_word.sounds_like,
'display_as': custom_word.display_as}
return self.request(method='PUT',
url=url.format(customization_id,
custom_word.word),
data=json.dumps(custom_word_fragment),
headers={'content-type': 'application/json'},
accept_json=True)
[docs] def list_custom_words(self, customization_id, word_type=None, sort=None):
url = '/v1/customizations/{0}/words'
qs = {}
if word_type:
if word_type in ['all', 'user', 'corpora']:
qs['word_type'] = word_type
else:
raise KeyError('word type must be all, user, or corpora')
if sort:
if sort in ['alphabetical', 'count']:
qs['sort'] = sort
else:
raise KeyError('sort must be alphabetical or count')
return self.request(method='GET',
url=url.format(customization_id),
params=qs,
accept_json=True)
[docs] def get_custom_word(self, customization_id, custom_word):
url = '/v1/customizations/{0}/words/{1}'
word = None
if isinstance(custom_word, str):
word = custom_word
else:
word = custom_word.word
return self.request(method='GET',
url=url.format(customization_id, word),
accept_json=True)
[docs] def delete_custom_word(self, customization_id, custom_word):
url = '/v1/customizations/{0}/words/{1}'
word = None
if isinstance(custom_word, str):
word = custom_word
else:
word = custom_word.word
return self.request(method='DELETE',
url=url.format(customization_id, word),
accept_json=True)