Blame - docs/dyn/speech_v1.speech.html - platform/external/python/google-api-python-client

2017-06-06 18:46:08 -0400

[diff] [blame]

100

# process the request.

101

# request.

Bu Sun Kim

2020-05-27 12:20:54 -0700

[diff] [blame^]

102

"enableWordTimeOffsets": True or False, # If `true`, the top result includes a list of words and

103

# the start and end time offsets (timestamps) for those words. If

104

# `false`, no word-level time offset information is returned. The default is

105

# `false`.

106

"diarizationConfig": { # Config to enable speaker diarization. # Config to enable speaker diarization and set additional

107

# parameters to make diarization better suited for your application.

108

# Note: When this is enabled, we send all the words from the beginning of the

109

# audio for the top alternative in every consecutive STREAMING responses.

110

# This is done in order to improve our speaker tags as our models learn to

111

# identify the speakers in the conversation over time.

112

# For non-streaming requests, the diarization results will be provided only

113

# in the top alternative of the FINAL SpeechRecognitionResult.

114

"minSpeakerCount": 42, # Minimum number of speakers in the conversation. This range gives you more

115

# flexibility by allowing the system to automatically determine the correct

116

# number of speakers. If not set, the default value is 2.

117

"maxSpeakerCount": 42, # Maximum number of speakers in the conversation. This range gives you more

118

# flexibility by allowing the system to automatically determine the correct

119

# number of speakers. If not set, the default value is 6.

120

"speakerTag": 42, # Output only. Unused.

121

"enableSpeakerDiarization": True or False, # If 'true', enables speaker detection for each recognized word in

122

# the top alternative of the recognition result using a speaker_tag provided

123

# in the WordInfo.

124

},

125

"languageCode": "A String", # Required. The language of the supplied audio as a

126

# [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) language tag.

127

# Example: "en-US".

128

# See [Language

129

# Support](https://cloud.google.com/speech-to-text/docs/languages) for a list

130

# of the currently supported language codes.

131

"profanityFilter": True or False, # If set to `true`, the server will attempt to filter out

132

# profanities, replacing all but the initial character in each filtered word

133

# with asterisks, e.g. "f***". If set to `false` or omitted, profanities

134

# won't be filtered out.

135

"useEnhanced": True or False, # Set to true to use an enhanced model for speech recognition.

136

# If `use_enhanced` is set to true and the `model` field is not set, then

137

# an appropriate enhanced model is chosen if an enhanced model exists for

138

# the audio.

139

#

140

# If `use_enhanced` is true and an enhanced version of the specified model

141

# does not exist, then the speech is recognized using the standard version

142

# of the specified model.

143

"metadata": { # Description of audio data to be recognized. # Metadata regarding this request.

144

"originalMediaType": "A String", # The original media the speech was recorded on.

145

"recordingDeviceType": "A String", # The type of device the speech was recorded with.

146

"interactionType": "A String", # The use case most closely describing the audio content to be recognized.

147

"audioTopic": "A String", # Description of the content. Eg. "Recordings of federal supreme court

148

# hearings from 2012".

149

"originalMimeType": "A String", # Mime type of the original audio file. For example `audio/m4a`,

150

# `audio/x-alaw-basic`, `audio/mp3`, `audio/3gpp`.

151

# A list of possible audio mime types is maintained at

152

# http://www.iana.org/assignments/media-types/media-types.xhtml#audio

153

"recordingDeviceName": "A String", # The device used to make the recording. Examples 'Nexus 5X' or

154

# 'Polycom SoundStation IP 6000' or 'POTS' or 'VoIP' or

155

# 'Cardioid Microphone'.

156

"industryNaicsCodeOfAudio": 42, # The industry vertical to which this speech recognition request most

157

# closely applies. This is most indicative of the topics contained

158

# in the audio. Use the 6-digit NAICS code to identify the industry

159

# vertical - see https://www.naics.com/search/.

160

"microphoneDistance": "A String", # The audio type that most closely describes the audio being recognized.

161

},

162

"sampleRateHertz": 42, # Sample rate in Hertz of the audio data sent in all

163

# `RecognitionAudio` messages. Valid values are: 8000-48000.

164

# 16000 is optimal. For best results, set the sampling rate of the audio

165

# source to 16000 Hz. If that's not possible, use the native sample rate of

166

# the audio source (instead of re-sampling).

167

# This field is optional for FLAC and WAV audio files, but is

168

# required for all other audio formats. For details, see AudioEncoding.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

169

"enableSeparateRecognitionPerChannel": True or False, # This needs to be set to `true` explicitly and `audio_channel_count` > 1

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

170

# to get each channel recognized separately. The recognition result will

171

# contain a `channel_tag` field to state which channel that result belongs

172

# to. If this is not true, we will only recognize the first channel. The

173

# request is billed cumulatively for all channels recognized:

174

# `audio_channel_count` multiplied by the length of the audio.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

175

"enableAutomaticPunctuation": True or False, # If 'true', adds punctuation to recognition result hypotheses.

176

# This feature is only available in select languages. Setting this for

177

# requests in other languages has no effect at all.

178

# The default 'false' value does not add punctuation to result hypotheses.

179

"maxAlternatives": 42, # Maximum number of recognition hypotheses to be returned.

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

180

# Specifically, the maximum number of `SpeechRecognitionAlternative` messages

181

# within each `SpeechRecognitionResult`.

182

# The server may return fewer than `max_alternatives`.

183

# Valid values are `0`-`30`. A value of `0` or `1` will return a maximum of

184

# one. If omitted, will return a maximum of one.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

185

"speechContexts": [ # Array of SpeechContext.

186

# A means to provide context to assist the speech recognition. For more

187

# information, see

188

# [speech

189

# adaptation](https://cloud.google.com/speech-to-text/docs/context-strength).

190

{ # Provides "hints" to the speech recognizer to favor specific words and phrases

191

# in the results.

192

"phrases": [ # A list of strings containing words and phrases "hints" so that

193

# the speech recognition is more likely to recognize them. This can be used

194

# to improve the accuracy for specific words and phrases, for example, if

195

# specific commands are typically spoken by the user. This can also be used

196

# to add additional words to the vocabulary of the recognizer. See

197

# [usage limits](https://cloud.google.com/speech-to-text/quotas#content).

198

#

199

# List items can also be set to classes for groups of words that represent

200

# common concepts that occur in natural language. For example, rather than

201

# providing phrase hints for every month of the year, using the $MONTH class

202

# improves the likelihood of correctly transcribing audio that includes

# months.

"A String",

],

},

],

Bu Sun Kim

2020-05-27 12:20:54 -0700

[diff] [blame^]

208

"encoding": "A String", # Encoding of audio data sent in all `RecognitionAudio` messages.

209

# This field is optional for `FLAC` and `WAV` audio files and required

210

# for all other audio formats. For details, see AudioEncoding.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

211

"model": "A String", # Which model to select for the given request. Select the model

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

212

# best suited to your domain to get best results. If a model is not

213

# explicitly specified, then we auto-select a model based on the parameters

214

# in the RecognitionConfig.

Dan O'Meara

2020-05-01 07:42:23 -0700

[diff] [blame]

215

# <table>

216

# <tr>

217

# <td><b>Model</b></td>

218

# <td><b>Description</b></td>

219

# </tr>

220

# <tr>

221

# <td><code>command_and_search</code></td>

222

# <td>Best for short queries such as voice commands or voice search.</td>

223

# </tr>

224

# <tr>

225

# <td><code>phone_call</code></td>

226

# <td>Best for audio that originated from a phone call (typically

227

# recorded at an 8khz sampling rate).</td>

228

# </tr>

229

# <tr>

230

# <td><code>video</code></td>

231

# <td>Best for audio that originated from from video or includes multiple

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

232

# speakers. Ideally the audio is recorded at a 16khz or greater

233

# sampling rate. This is a premium model that costs more than the

Dan O'Meara

2020-05-01 07:42:23 -0700

[diff] [blame]

234

# standard rate.</td>

235

# </tr>

236

# <tr>

237

# <td><code>default</code></td>

238

# <td>Best for audio that is not one of the specific audio models.

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

239

# For example, long-form audio. Ideally the audio is high-fidelity,

Dan O'Meara

2020-05-01 07:42:23 -0700

[diff] [blame]

240

# recorded at a 16khz or greater sampling rate.</td>

241

# </tr>

242

# </table>

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

243

"audioChannelCount": 42, # The number of channels in the input audio data.

244

# ONLY set this for MULTI-CHANNEL recognition.

245

# Valid values for LINEAR16 and FLAC are `1`-`8`.

246

# Valid values for OGG_OPUS are '1'-'254'.

247

# Valid value for MULAW, AMR, AMR_WB and SPEEX_WITH_HEADER_BYTE is only `1`.

248

# If `0` or omitted, defaults to one channel (mono).

249

# Note: We only recognize the first channel by default.

250

# To perform independent recognition on each channel set

251

# `enable_separate_recognition_per_channel` to 'true'.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

252

},

253

"audio": { # Contains audio data in the encoding specified in the `RecognitionConfig`. # Required. The audio data to be recognized.

254

# Either `content` or `uri` must be supplied. Supplying both or neither

255

# returns google.rpc.Code.INVALID_ARGUMENT. See

256

# [content limits](https://cloud.google.com/speech-to-text/quotas#content).

Bu Sun Kim

2020-05-27 12:20:54 -0700

[diff] [blame^]

257

"content": "A String", # The audio data bytes encoded as specified in

258

# `RecognitionConfig`. Note: as with all bytes fields, proto buffers use a

259

# pure binary representation, whereas JSON representations use base64.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

260

"uri": "A String", # URI that points to a file that contains audio data bytes as specified in

261

# `RecognitionConfig`. The file must not be compressed (for example, gzip).

262

# Currently, only Google Cloud Storage URIs are

263

# supported, which must be specified in the following format:

264

# `gs://bucket_name/object_name` (other URI formats return

265

# google.rpc.Code.INVALID_ARGUMENT). For more information, see

266

# [Request URIs](https://cloud.google.com/storage/docs/reference-uris).

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

},

}

x__xgafv: string, V1 error format.

Allowed values

1 - v1 error format

2 - v2 error format

Returns:

An object of the form:

277

278

{ # This resource represents a long-running operation that is the result of a

279

# network API call.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

280

"response": { # The normal response of the operation in case of success. If the original

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

281

# method returns no data on success, such as `Delete`, the response is

282

# `google.protobuf.Empty`. If the original method is standard

283

# `Get`/`Create`/`Update`, the response should be the resource. For other

284

# methods, the response should have the type `XxxResponse`, where `Xxx`

285

# is the original method name. For example, if the original method name

286

# is `TakeSnapshot()`, the inferred response type is

287

# `TakeSnapshotResponse`.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

288

"a_key": "", # Properties of the object. Contains field @type with type URL.

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

289

},

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

290

"name": "A String", # The server-assigned name, which is only unique within the same service that

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

291

# originally returns it. If you use the default HTTP mapping, the

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

292

# `name` should be a resource name ending with `operations/{unique_id}`.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

293

"error": { # The `Status` type defines a logical error model that is suitable for # The error result of the operation in case of failure or cancellation.

294

# different programming environments, including REST APIs and RPC APIs. It is

295

# used by [gRPC](https://github.com/grpc). Each `Status` message contains

296

# three pieces of data: error code, error message, and error details.

297

#

298

# You can find out more about this error model and how to work with it in the

299

# [API Design Guide](https://cloud.google.com/apis/design/errors).

300

"details": [ # A list of messages that carry the error details. There is a common set of

301

# message types for APIs to use.

302

{

303

"a_key": "", # Properties of the object. Contains field @type with type URL.

304

},

305

],

306

"code": 42, # The status code, which should be an enum value of google.rpc.Code.

307

"message": "A String", # A developer-facing error message, which should be in English. Any

308

# user-facing error message should be localized and sent in the

309

# google.rpc.Status.details field, or localized by the client.

310

},

311

"metadata": { # Service-specific metadata associated with the operation. It typically

312

# contains progress information and common metadata such as create time.

313

# Some services might not provide such metadata. Any method that returns a

314

# long-running operation should document the metadata type, if any.

315

"a_key": "", # Properties of the object. Contains field @type with type URL.

316

},

Bu Sun Kim

2020-05-27 12:20:54 -0700

[diff] [blame^]

317

"done": True or False, # If the value is `false`, it means the operation is still in progress.

318

# If `true`, the operation is completed, and either `error` or `response` is

319

# available.

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

}</pre>

</div>

Dan O'Meara

2020-05-01 07:42:23 -0700

[diff] [blame]

324

<code class="details" id="recognize">recognize(body=None, x__xgafv=None)</code>

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

325

<pre>Performs synchronous speech recognition: receive results after all audio

326

has been sent and processed.

327

328

Args:

Dan O'Meara

2020-05-01 07:42:23 -0700

[diff] [blame]

329

body: object, The request body.

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

330

The object takes the form of:

331

332

{ # The top-level message sent by the client for the `Recognize` method.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

333

"config": { # Provides information to the recognizer that specifies how to process the # Required. Provides information to the recognizer that specifies how to

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

334

# process the request.

335

# request.

Bu Sun Kim

2020-05-27 12:20:54 -0700

[diff] [blame^]

336

"enableWordTimeOffsets": True or False, # If `true`, the top result includes a list of words and

337

# the start and end time offsets (timestamps) for those words. If

338

# `false`, no word-level time offset information is returned. The default is

339

# `false`.

340

"diarizationConfig": { # Config to enable speaker diarization. # Config to enable speaker diarization and set additional

341

# parameters to make diarization better suited for your application.

342

# Note: When this is enabled, we send all the words from the beginning of the

343

# audio for the top alternative in every consecutive STREAMING responses.

344

# This is done in order to improve our speaker tags as our models learn to

345

# identify the speakers in the conversation over time.

346

# For non-streaming requests, the diarization results will be provided only

347

# in the top alternative of the FINAL SpeechRecognitionResult.

348

"minSpeakerCount": 42, # Minimum number of speakers in the conversation. This range gives you more

349

# flexibility by allowing the system to automatically determine the correct

350

# number of speakers. If not set, the default value is 2.

351

"maxSpeakerCount": 42, # Maximum number of speakers in the conversation. This range gives you more

352

# flexibility by allowing the system to automatically determine the correct

353

# number of speakers. If not set, the default value is 6.

354

"speakerTag": 42, # Output only. Unused.

355

"enableSpeakerDiarization": True or False, # If 'true', enables speaker detection for each recognized word in

356

# the top alternative of the recognition result using a speaker_tag provided

357

# in the WordInfo.

358

},

359

"languageCode": "A String", # Required. The language of the supplied audio as a

360

# [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) language tag.

361

# Example: "en-US".

362

# See [Language

363

# Support](https://cloud.google.com/speech-to-text/docs/languages) for a list

364

# of the currently supported language codes.

365

"profanityFilter": True or False, # If set to `true`, the server will attempt to filter out

366

# profanities, replacing all but the initial character in each filtered word

367

# with asterisks, e.g. "f***". If set to `false` or omitted, profanities

368

# won't be filtered out.

369

"useEnhanced": True or False, # Set to true to use an enhanced model for speech recognition.

370

# If `use_enhanced` is set to true and the `model` field is not set, then

371

# an appropriate enhanced model is chosen if an enhanced model exists for

372

# the audio.

373

#

374

# If `use_enhanced` is true and an enhanced version of the specified model

375

# does not exist, then the speech is recognized using the standard version

376

# of the specified model.

377

"metadata": { # Description of audio data to be recognized. # Metadata regarding this request.

378

"originalMediaType": "A String", # The original media the speech was recorded on.

379

"recordingDeviceType": "A String", # The type of device the speech was recorded with.

380

"interactionType": "A String", # The use case most closely describing the audio content to be recognized.

381

"audioTopic": "A String", # Description of the content. Eg. "Recordings of federal supreme court

382

# hearings from 2012".

383

"originalMimeType": "A String", # Mime type of the original audio file. For example `audio/m4a`,

384

# `audio/x-alaw-basic`, `audio/mp3`, `audio/3gpp`.

385

# A list of possible audio mime types is maintained at

386

# http://www.iana.org/assignments/media-types/media-types.xhtml#audio

387

"recordingDeviceName": "A String", # The device used to make the recording. Examples 'Nexus 5X' or

388

# 'Polycom SoundStation IP 6000' or 'POTS' or 'VoIP' or

389

# 'Cardioid Microphone'.

390

"industryNaicsCodeOfAudio": 42, # The industry vertical to which this speech recognition request most

391

# closely applies. This is most indicative of the topics contained

392

# in the audio. Use the 6-digit NAICS code to identify the industry

393

# vertical - see https://www.naics.com/search/.

394

"microphoneDistance": "A String", # The audio type that most closely describes the audio being recognized.

395

},

396

"sampleRateHertz": 42, # Sample rate in Hertz of the audio data sent in all

397

# `RecognitionAudio` messages. Valid values are: 8000-48000.

398

# 16000 is optimal. For best results, set the sampling rate of the audio

399

# source to 16000 Hz. If that's not possible, use the native sample rate of

400

# the audio source (instead of re-sampling).

401

# This field is optional for FLAC and WAV audio files, but is

402

# required for all other audio formats. For details, see AudioEncoding.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

403

"enableSeparateRecognitionPerChannel": True or False, # This needs to be set to `true` explicitly and `audio_channel_count` > 1

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

404

# to get each channel recognized separately. The recognition result will

405

# contain a `channel_tag` field to state which channel that result belongs

406

# to. If this is not true, we will only recognize the first channel. The

407

# request is billed cumulatively for all channels recognized:

408

# `audio_channel_count` multiplied by the length of the audio.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

409

"enableAutomaticPunctuation": True or False, # If 'true', adds punctuation to recognition result hypotheses.

410

# This feature is only available in select languages. Setting this for

411

# requests in other languages has no effect at all.

412

# The default 'false' value does not add punctuation to result hypotheses.

413

"maxAlternatives": 42, # Maximum number of recognition hypotheses to be returned.

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

414

# Specifically, the maximum number of `SpeechRecognitionAlternative` messages

415

# within each `SpeechRecognitionResult`.

416

# The server may return fewer than `max_alternatives`.

417

# Valid values are `0`-`30`. A value of `0` or `1` will return a maximum of

418

# one. If omitted, will return a maximum of one.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

419

"speechContexts": [ # Array of SpeechContext.

420

# A means to provide context to assist the speech recognition. For more

421

# information, see

422

# [speech

423

# adaptation](https://cloud.google.com/speech-to-text/docs/context-strength).

424

{ # Provides "hints" to the speech recognizer to favor specific words and phrases

425

# in the results.

426

"phrases": [ # A list of strings containing words and phrases "hints" so that

427

# the speech recognition is more likely to recognize them. This can be used

428

# to improve the accuracy for specific words and phrases, for example, if

429

# specific commands are typically spoken by the user. This can also be used

430

# to add additional words to the vocabulary of the recognizer. See

431

# [usage limits](https://cloud.google.com/speech-to-text/quotas#content).

432

#

433

# List items can also be set to classes for groups of words that represent

434

# common concepts that occur in natural language. For example, rather than

435

# providing phrase hints for every month of the year, using the $MONTH class

436

# improves the likelihood of correctly transcribing audio that includes

# months.

"A String",

],

},

],

Bu Sun Kim

2020-05-27 12:20:54 -0700

[diff] [blame^]

442

"encoding": "A String", # Encoding of audio data sent in all `RecognitionAudio` messages.

443

# This field is optional for `FLAC` and `WAV` audio files and required

444

# for all other audio formats. For details, see AudioEncoding.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

445

"model": "A String", # Which model to select for the given request. Select the model

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

446

# best suited to your domain to get best results. If a model is not

447

# explicitly specified, then we auto-select a model based on the parameters

448

# in the RecognitionConfig.

Dan O'Meara

2020-05-01 07:42:23 -0700

[diff] [blame]

449

# <table>

450

# <tr>

451

# <td><b>Model</b></td>

452

# <td><b>Description</b></td>

453

# </tr>

454

# <tr>

455

# <td><code>command_and_search</code></td>

456

# <td>Best for short queries such as voice commands or voice search.</td>

457

# </tr>

458

# <tr>

459

# <td><code>phone_call</code></td>

460

# <td>Best for audio that originated from a phone call (typically

461

# recorded at an 8khz sampling rate).</td>

462

# </tr>

463

# <tr>

464

# <td><code>video</code></td>

465

# <td>Best for audio that originated from from video or includes multiple

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

466

# speakers. Ideally the audio is recorded at a 16khz or greater

467

# sampling rate. This is a premium model that costs more than the

Dan O'Meara

2020-05-01 07:42:23 -0700

[diff] [blame]

468

# standard rate.</td>

469

# </tr>

470

# <tr>

471

# <td><code>default</code></td>

472

# <td>Best for audio that is not one of the specific audio models.

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

473

# For example, long-form audio. Ideally the audio is high-fidelity,

Dan O'Meara

2020-05-01 07:42:23 -0700

[diff] [blame]

474

# recorded at a 16khz or greater sampling rate.</td>

475

# </tr>

476

# </table>

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

477

"audioChannelCount": 42, # The number of channels in the input audio data.

478

# ONLY set this for MULTI-CHANNEL recognition.

479

# Valid values for LINEAR16 and FLAC are `1`-`8`.

480

# Valid values for OGG_OPUS are '1'-'254'.

481

# Valid value for MULAW, AMR, AMR_WB and SPEEX_WITH_HEADER_BYTE is only `1`.

482

# If `0` or omitted, defaults to one channel (mono).

483

# Note: We only recognize the first channel by default.

484

# To perform independent recognition on each channel set

485

# `enable_separate_recognition_per_channel` to 'true'.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

486

},

487

"audio": { # Contains audio data in the encoding specified in the `RecognitionConfig`. # Required. The audio data to be recognized.

488

# Either `content` or `uri` must be supplied. Supplying both or neither

489

# returns google.rpc.Code.INVALID_ARGUMENT. See

490

# [content limits](https://cloud.google.com/speech-to-text/quotas#content).

Bu Sun Kim

2020-05-27 12:20:54 -0700

[diff] [blame^]

491

"content": "A String", # The audio data bytes encoded as specified in

492

# `RecognitionConfig`. Note: as with all bytes fields, proto buffers use a

493

# pure binary representation, whereas JSON representations use base64.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

494

"uri": "A String", # URI that points to a file that contains audio data bytes as specified in

495

# `RecognitionConfig`. The file must not be compressed (for example, gzip).

496

# Currently, only Google Cloud Storage URIs are

497

# supported, which must be specified in the following format:

498

# `gs://bucket_name/object_name` (other URI formats return

499

# google.rpc.Code.INVALID_ARGUMENT). For more information, see

500

# [Request URIs](https://cloud.google.com/storage/docs/reference-uris).

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

},

}

x__xgafv: string, V1 error format.

Allowed values

1 - v1 error format

2 - v2 error format

Returns:

An object of the form:

511

512

{ # The only message returned to the client by the `Recognize` method. It

513

# contains the result as zero or more sequential `SpeechRecognitionResult`

514

# messages.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

515

"results": [ # Sequential list of transcription results corresponding to

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

516

# sequential portions of audio.

517

{ # A speech recognition result corresponding to a portion of the audio.

Bu Sun Kim

2020-05-27 12:20:54 -0700

[diff] [blame^]

518

"channelTag": 42, # For multi-channel audio, this is the channel number corresponding to the

519

# recognized result for the audio from that channel.

520

# For audio_channel_count = N, its output values can range from '1' to 'N'.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

521

"alternatives": [ # May contain one or more recognition hypotheses (up to the

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

522

# maximum specified in `max_alternatives`).

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

523

# These alternatives are ordered in terms of accuracy, with the top (first)

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

524

# alternative being the most probable, as ranked by the recognizer.

525

{ # Alternative hypotheses (a.k.a. n-best list).

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

526

"confidence": 3.14, # The confidence estimate between 0.0 and 1.0. A higher number

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

527

# indicates an estimated greater likelihood that the recognized words are

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

528

# correct. This field is set only for the top alternative of a non-streaming

529

# result or, of a streaming result where `is_final=true`.

530

# This field is not guaranteed to be accurate and users should not rely on it

531

# to be always provided.

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

532

# The default of 0.0 is a sentinel value indicating `confidence` was not set.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

533

"transcript": "A String", # Transcript text representing the words that the user spoke.

534

"words": [ # A list of word-specific information for each recognized word.

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

535

# Note: When `enable_speaker_diarization` is true, you will see all the words

536

# from the beginning of the audio.

537

{ # Word-specific information for recognized words.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

538

"endTime": "A String", # Time offset relative to the beginning of the audio,

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

539

# and corresponding to the end of the spoken word.

540

# This field is only set if `enable_word_time_offsets=true` and only

541

# in the top hypothesis.

542

# This is an experimental feature and the accuracy of the time offset can

543

# vary.

Bu Sun Kim

2020-05-20 12:08:20 -0700

[diff] [blame]

544

"startTime": "A String", # Time offset relative to the beginning of the audio,

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

545

# and corresponding to the start of the spoken word.

546

# This field is only set if `enable_word_time_offsets=true` and only

547

# in the top hypothesis.

548

# This is an experimental feature and the accuracy of the time offset can

549

# vary.

Bu Sun Kim

2020-05-27 12:20:54 -0700

[diff] [blame^]

550

"word": "A String", # The word corresponding to this set of information.

551

"speakerTag": 42, # Output only. A distinct integer value is assigned for every speaker within

552

# the audio. This field specifies which one of those speakers was detected to

553

# have spoken this word. Value ranges from '1' to diarization_speaker_count.

554

# speaker_tag is set if enable_speaker_diarization = 'true' and only in the

555

# top alternative.

Bu Sun Kim

2019-06-14 16:50:42 -0700

[diff] [blame]

556

},

557

],

Sai Cheemalapati

2017-06-06 18:46:08 -0400

[diff] [blame]

558

},

559

],

Sai Cheemalapati