1<html><body> 2<style> 3 4body, h1, h2, h3, div, span, p, pre, a { 5 margin: 0; 6 padding: 0; 7 border: 0; 8 font-weight: inherit; 9 font-style: inherit; 10 font-size: 100%; 11 font-family: inherit; 12 vertical-align: baseline; 13} 14 15body { 16 font-size: 13px; 17 padding: 1em; 18} 19 20h1 { 21 font-size: 26px; 22 margin-bottom: 1em; 23} 24 25h2 { 26 font-size: 24px; 27 margin-bottom: 1em; 28} 29 30h3 { 31 font-size: 20px; 32 margin-bottom: 1em; 33 margin-top: 1em; 34} 35 36pre, code { 37 line-height: 1.5; 38 font-family: Monaco, 'DejaVu Sans Mono', 'Bitstream Vera Sans Mono', 'Lucida Console', monospace; 39} 40 41pre { 42 margin-top: 0.5em; 43} 44 45h1, h2, h3, p { 46 font-family: Arial, sans serif; 47} 48 49h1, h2, h3 { 50 border-bottom: solid #CCC 1px; 51} 52 53.toc_element { 54 margin-top: 0.5em; 55} 56 57.firstline { 58 margin-left: 2 em; 59} 60 61.method { 62 margin-top: 1em; 63 border: solid 1px #CCC; 64 padding: 1em; 65 background: #EEE; 66} 67 68.details { 69 font-weight: bold; 70 font-size: 14px; 71} 72 73</style> 74 75<h1><a href="speech_v1p1beta1.html">Cloud Speech-to-Text API</a> . <a href="speech_v1p1beta1.speech.html">speech</a></h1> 76<h2>Instance Methods</h2> 77<p class="toc_element"> 78 <code><a href="#longrunningrecognize">longrunningrecognize(body, x__xgafv=None)</a></code></p> 79<p class="firstline">Performs asynchronous speech recognition: receive results via the</p> 80<p class="toc_element"> 81 <code><a href="#recognize">recognize(body, x__xgafv=None)</a></code></p> 82<p class="firstline">Performs synchronous speech recognition: receive results after all audio</p> 83<h3>Method Details</h3> 84<div class="method"> 85 <code class="details" id="longrunningrecognize">longrunningrecognize(body, x__xgafv=None)</code> 86 <pre>Performs asynchronous speech recognition: receive results via the 87google.longrunning.Operations interface. Returns either an 88`Operation.error` or an `Operation.response` which contains 89a `LongRunningRecognizeResponse` message. 90For more information on asynchronous speech recognition, see the 91[how-to](https://cloud.google.com/speech-to-text/docs/async-recognize). 92 93Args: 94 body: object, The request body. (required) 95 The object takes the form of: 96 97{ # The top-level message sent by the client for the `LongRunningRecognize` 98 # method. 99 "audio": { # Contains audio data in the encoding specified in the `RecognitionConfig`. # *Required* The audio data to be recognized. 100 # Either `content` or `uri` must be supplied. Supplying both or neither 101 # returns google.rpc.Code.INVALID_ARGUMENT. See 102 # [content limits](/speech-to-text/quotas#content). 103 "content": "A String", # The audio data bytes encoded as specified in 104 # `RecognitionConfig`. Note: as with all bytes fields, proto buffers use a 105 # pure binary representation, whereas JSON representations use base64. 106 "uri": "A String", # URI that points to a file that contains audio data bytes as specified in 107 # `RecognitionConfig`. The file must not be compressed (for example, gzip). 108 # Currently, only Google Cloud Storage URIs are 109 # supported, which must be specified in the following format: 110 # `gs://bucket_name/object_name` (other URI formats return 111 # google.rpc.Code.INVALID_ARGUMENT). For more information, see 112 # [Request URIs](https://cloud.google.com/storage/docs/reference-uris). 113 }, 114 "config": { # Provides information to the recognizer that specifies how to process the # *Required* Provides information to the recognizer that specifies how to 115 # process the request. 116 # request. 117 "languageCode": "A String", # *Required* The language of the supplied audio as a 118 # [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) language tag. 119 # Example: "en-US". 120 # See [Language Support](/speech-to-text/docs/languages) 121 # for a list of the currently supported language codes. 122 "audioChannelCount": 42, # *Optional* The number of channels in the input audio data. 123 # ONLY set this for MULTI-CHANNEL recognition. 124 # Valid values for LINEAR16 and FLAC are `1`-`8`. 125 # Valid values for OGG_OPUS are '1'-'254'. 126 # Valid value for MULAW, AMR, AMR_WB and SPEEX_WITH_HEADER_BYTE is only `1`. 127 # If `0` or omitted, defaults to one channel (mono). 128 # Note: We only recognize the first channel by default. 129 # To perform independent recognition on each channel set 130 # `enable_separate_recognition_per_channel` to 'true'. 131 "encoding": "A String", # Encoding of audio data sent in all `RecognitionAudio` messages. 132 # This field is optional for `FLAC` and `WAV` audio files and required 133 # for all other audio formats. For details, see AudioEncoding. 134 "enableAutomaticPunctuation": True or False, # *Optional* If 'true', adds punctuation to recognition result hypotheses. 135 # This feature is only available in select languages. Setting this for 136 # requests in other languages has no effect at all. 137 # The default 'false' value does not add punctuation to result hypotheses. 138 # Note: This is currently offered as an experimental service, complimentary 139 # to all users. In the future this may be exclusively available as a 140 # premium feature. 141 "alternativeLanguageCodes": [ # *Optional* A list of up to 3 additional 142 # [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) language tags, 143 # listing possible alternative languages of the supplied audio. 144 # See [Language Support](/speech-to-text/docs/languages) 145 # for a list of the currently supported language codes. 146 # If alternative languages are listed, recognition result will contain 147 # recognition in the most likely language detected including the main 148 # language_code. The recognition result will include the language tag 149 # of the language detected in the audio. 150 # Note: This feature is only supported for Voice Command and Voice Search 151 # use cases and performance may vary for other use cases (e.g., phone call 152 # transcription). 153 "A String", 154 ], 155 "enableSeparateRecognitionPerChannel": True or False, # This needs to be set to `true` explicitly and `audio_channel_count` > 1 156 # to get each channel recognized separately. The recognition result will 157 # contain a `channel_tag` field to state which channel that result belongs 158 # to. If this is not true, we will only recognize the first channel. The 159 # request is billed cumulatively for all channels recognized: 160 # `audio_channel_count` multiplied by the length of the audio. 161 "enableWordTimeOffsets": True or False, # *Optional* If `true`, the top result includes a list of words and 162 # the start and end time offsets (timestamps) for those words. If 163 # `false`, no word-level time offset information is returned. The default is 164 # `false`. 165 "enableSpeakerDiarization": True or False, # *Optional* If 'true', enables speaker detection for each recognized word in 166 # the top alternative of the recognition result using a speaker_tag provided 167 # in the WordInfo. 168 # Note: Use diarization_config instead. This field will be DEPRECATED soon. 169 "maxAlternatives": 42, # *Optional* Maximum number of recognition hypotheses to be returned. 170 # Specifically, the maximum number of `SpeechRecognitionAlternative` messages 171 # within each `SpeechRecognitionResult`. 172 # The server may return fewer than `max_alternatives`. 173 # Valid values are `0`-`30`. A value of `0` or `1` will return a maximum of 174 # one. If omitted, will return a maximum of one. 175 "profanityFilter": True or False, # *Optional* If set to `true`, the server will attempt to filter out 176 # profanities, replacing all but the initial character in each filtered word 177 # with asterisks, e.g. "f***". If set to `false` or omitted, profanities 178 # won't be filtered out. 179 "useEnhanced": True or False, # *Optional* Set to true to use an enhanced model for speech recognition. 180 # If `use_enhanced` is set to true and the `model` field is not set, then 181 # an appropriate enhanced model is chosen if: 182 # 1. project is eligible for requesting enhanced models 183 # 2. an enhanced model exists for the audio 184 # 185 # If `use_enhanced` is true and an enhanced version of the specified model 186 # does not exist, then the speech is recognized using the standard version 187 # of the specified model. 188 # 189 # Enhanced speech models require that you opt-in to data logging using 190 # instructions in the 191 # [documentation](/speech-to-text/docs/enable-data-logging). If you set 192 # `use_enhanced` to true and you have not enabled audio logging, then you 193 # will receive an error. 194 "sampleRateHertz": 42, # Sample rate in Hertz of the audio data sent in all 195 # `RecognitionAudio` messages. Valid values are: 8000-48000. 196 # 16000 is optimal. For best results, set the sampling rate of the audio 197 # source to 16000 Hz. If that's not possible, use the native sample rate of 198 # the audio source (instead of re-sampling). 199 # This field is optional for FLAC and WAV audio files, but is 200 # required for all other audio formats. For details, see AudioEncoding. 201 "diarizationSpeakerCount": 42, # *Optional* 202 # If set, specifies the estimated number of speakers in the conversation. 203 # If not set, defaults to '2'. 204 # Ignored unless enable_speaker_diarization is set to true." 205 # Note: Use diarization_config instead. This field will be DEPRECATED soon. 206 "enableWordConfidence": True or False, # *Optional* If `true`, the top result includes a list of words and the 207 # confidence for those words. If `false`, no word-level confidence 208 # information is returned. The default is `false`. 209 "model": "A String", # *Optional* Which model to select for the given request. Select the model 210 # best suited to your domain to get best results. If a model is not 211 # explicitly specified, then we auto-select a model based on the parameters 212 # in the RecognitionConfig. 213 # <table> 214 # <tr> 215 # <td><b>Model</b></td> 216 # <td><b>Description</b></td> 217 # </tr> 218 # <tr> 219 # <td><code>command_and_search</code></td> 220 # <td>Best for short queries such as voice commands or voice search.</td> 221 # </tr> 222 # <tr> 223 # <td><code>phone_call</code></td> 224 # <td>Best for audio that originated from a phone call (typically 225 # recorded at an 8khz sampling rate).</td> 226 # </tr> 227 # <tr> 228 # <td><code>video</code></td> 229 # <td>Best for audio that originated from from video or includes multiple 230 # speakers. Ideally the audio is recorded at a 16khz or greater 231 # sampling rate. This is a premium model that costs more than the 232 # standard rate.</td> 233 # </tr> 234 # <tr> 235 # <td><code>default</code></td> 236 # <td>Best for audio that is not one of the specific audio models. 237 # For example, long-form audio. Ideally the audio is high-fidelity, 238 # recorded at a 16khz or greater sampling rate.</td> 239 # </tr> 240 # </table> 241 "diarizationConfig": { # *Optional* Config to enable speaker diarization and set additional 242 # parameters to make diarization better suited for your application. 243 # Note: When this is enabled, we send all the words from the beginning of the 244 # audio for the top alternative in every consecutive STREAMING responses. 245 # This is done in order to improve our speaker tags as our models learn to 246 # identify the speakers in the conversation over time. 247 # For non-streaming requests, the diarization results will be provided only 248 # in the top alternative of the FINAL SpeechRecognitionResult. 249 "minSpeakerCount": 42, # *Optional* Only used if diarization_speaker_count is not set. 250 # Minimum number of speakers in the conversation. This range gives you more 251 # flexibility by allowing the system to automatically determine the correct 252 # number of speakers. If not set, the default value is 2. 253 "enableSpeakerDiarization": True or False, # *Optional* If 'true', enables speaker detection for each recognized word in 254 # the top alternative of the recognition result using a speaker_tag provided 255 # in the WordInfo. 256 "maxSpeakerCount": 42, # *Optional* Only used if diarization_speaker_count is not set. 257 # Maximum number of speakers in the conversation. This range gives you more 258 # flexibility by allowing the system to automatically determine the correct 259 # number of speakers. If not set, the default value is 6. 260 }, 261 "speechContexts": [ # *Optional* array of SpeechContext. 262 # A means to provide context to assist the speech recognition. For more 263 # information, see [Phrase Hints](/speech-to-text/docs/basics#phrase-hints). 264 { # Provides "hints" to the speech recognizer to favor specific words and phrases 265 # in the results. 266 "phrases": [ # *Optional* A list of strings containing words and phrases "hints" so that 267 # the speech recognition is more likely to recognize them. This can be used 268 # to improve the accuracy for specific words and phrases, for example, if 269 # specific commands are typically spoken by the user. This can also be used 270 # to add additional words to the vocabulary of the recognizer. See 271 # [usage limits](/speech-to-text/quotas#content). 272 # 273 # List items can also be set to classes for groups of words that represent 274 # common concepts that occur in natural language. For example, rather than 275 # providing phrase hints for every month of the year, using the $MONTH class 276 # improves the likelihood of correctly transcribing audio that includes 277 # months. 278 "A String", 279 ], 280 "boost": 3.14, # Hint Boost. Positive value will increase the probability that a specific 281 # phrase will be recognized over other similar sounding phrases. The higher 282 # the boost, the higher the chance of false positive recognition as well. 283 # Negative boost values would correspond to anti-biasing. Anti-biasing is not 284 # enabled, so negative boost will simply be ignored. Though `boost` can 285 # accept a wide range of positive values, most use cases are best served with 286 # values between 0 and 20. We recommend using a binary search approach to 287 # finding the optimal value for your use case. 288 }, 289 ], 290 "metadata": { # Description of audio data to be recognized. # *Optional* Metadata regarding this request. 291 "recordingDeviceType": "A String", # The type of device the speech was recorded with. 292 "originalMediaType": "A String", # The original media the speech was recorded on. 293 "microphoneDistance": "A String", # The audio type that most closely describes the audio being recognized. 294 "obfuscatedId": "A String", # Obfuscated (privacy-protected) ID of the user, to identify number of 295 # unique users using the service. 296 "recordingDeviceName": "A String", # The device used to make the recording. Examples 'Nexus 5X' or 297 # 'Polycom SoundStation IP 6000' or 'POTS' or 'VoIP' or 298 # 'Cardioid Microphone'. 299 "industryNaicsCodeOfAudio": 42, # The industry vertical to which this speech recognition request most 300 # closely applies. This is most indicative of the topics contained 301 # in the audio. Use the 6-digit NAICS code to identify the industry 302 # vertical - see https://www.naics.com/search/. 303 "audioTopic": "A String", # Description of the content. Eg. "Recordings of federal supreme court 304 # hearings from 2012". 305 "originalMimeType": "A String", # Mime type of the original audio file. For example `audio/m4a`, 306 # `audio/x-alaw-basic`, `audio/mp3`, `audio/3gpp`. 307 # A list of possible audio mime types is maintained at 308 # http://www.iana.org/assignments/media-types/media-types.xhtml#audio 309 "interactionType": "A String", # The use case most closely describing the audio content to be recognized. 310 }, 311 }, 312 } 313 314 x__xgafv: string, V1 error format. 315 Allowed values 316 1 - v1 error format 317 2 - v2 error format 318 319Returns: 320 An object of the form: 321 322 { # This resource represents a long-running operation that is the result of a 323 # network API call. 324 "metadata": { # Service-specific metadata associated with the operation. It typically 325 # contains progress information and common metadata such as create time. 326 # Some services might not provide such metadata. Any method that returns a 327 # long-running operation should document the metadata type, if any. 328 "a_key": "", # Properties of the object. Contains field @type with type URL. 329 }, 330 "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. 331 # different programming environments, including REST APIs and RPC APIs. It is 332 # used by [gRPC](https://github.com/grpc). Each `Status` message contains 333 # three pieces of data: error code, error message, and error details. 334 # 335 # You can find out more about this error model and how to work with it in the 336 # [API Design Guide](https://cloud.google.com/apis/design/errors). 337 "message": "A String", # A developer-facing error message, which should be in English. Any 338 # user-facing error message should be localized and sent in the 339 # google.rpc.Status.details field, or localized by the client. 340 "code": 42, # The status code, which should be an enum value of google.rpc.Code. 341 "details": [ # A list of messages that carry the error details. There is a common set of 342 # message types for APIs to use. 343 { 344 "a_key": "", # Properties of the object. Contains field @type with type URL. 345 }, 346 ], 347 }, 348 "done": True or False, # If the value is `false`, it means the operation is still in progress. 349 # If `true`, the operation is completed, and either `error` or `response` is 350 # available. 351 "response": { # The normal response of the operation in case of success. If the original 352 # method returns no data on success, such as `Delete`, the response is 353 # `google.protobuf.Empty`. If the original method is standard 354 # `Get`/`Create`/`Update`, the response should be the resource. For other 355 # methods, the response should have the type `XxxResponse`, where `Xxx` 356 # is the original method name. For example, if the original method name 357 # is `TakeSnapshot()`, the inferred response type is 358 # `TakeSnapshotResponse`. 359 "a_key": "", # Properties of the object. Contains field @type with type URL. 360 }, 361 "name": "A String", # The server-assigned name, which is only unique within the same service that 362 # originally returns it. If you use the default HTTP mapping, the 363 # `name` should be a resource name ending with `operations/{unique_id}`. 364 }</pre> 365</div> 366 367<div class="method"> 368 <code class="details" id="recognize">recognize(body, x__xgafv=None)</code> 369 <pre>Performs synchronous speech recognition: receive results after all audio 370has been sent and processed. 371 372Args: 373 body: object, The request body. (required) 374 The object takes the form of: 375 376{ # The top-level message sent by the client for the `Recognize` method. 377 "audio": { # Contains audio data in the encoding specified in the `RecognitionConfig`. # *Required* The audio data to be recognized. 378 # Either `content` or `uri` must be supplied. Supplying both or neither 379 # returns google.rpc.Code.INVALID_ARGUMENT. See 380 # [content limits](/speech-to-text/quotas#content). 381 "content": "A String", # The audio data bytes encoded as specified in 382 # `RecognitionConfig`. Note: as with all bytes fields, proto buffers use a 383 # pure binary representation, whereas JSON representations use base64. 384 "uri": "A String", # URI that points to a file that contains audio data bytes as specified in 385 # `RecognitionConfig`. The file must not be compressed (for example, gzip). 386 # Currently, only Google Cloud Storage URIs are 387 # supported, which must be specified in the following format: 388 # `gs://bucket_name/object_name` (other URI formats return 389 # google.rpc.Code.INVALID_ARGUMENT). For more information, see 390 # [Request URIs](https://cloud.google.com/storage/docs/reference-uris). 391 }, 392 "config": { # Provides information to the recognizer that specifies how to process the # *Required* Provides information to the recognizer that specifies how to 393 # process the request. 394 # request. 395 "languageCode": "A String", # *Required* The language of the supplied audio as a 396 # [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) language tag. 397 # Example: "en-US". 398 # See [Language Support](/speech-to-text/docs/languages) 399 # for a list of the currently supported language codes. 400 "audioChannelCount": 42, # *Optional* The number of channels in the input audio data. 401 # ONLY set this for MULTI-CHANNEL recognition. 402 # Valid values for LINEAR16 and FLAC are `1`-`8`. 403 # Valid values for OGG_OPUS are '1'-'254'. 404 # Valid value for MULAW, AMR, AMR_WB and SPEEX_WITH_HEADER_BYTE is only `1`. 405 # If `0` or omitted, defaults to one channel (mono). 406 # Note: We only recognize the first channel by default. 407 # To perform independent recognition on each channel set 408 # `enable_separate_recognition_per_channel` to 'true'. 409 "encoding": "A String", # Encoding of audio data sent in all `RecognitionAudio` messages. 410 # This field is optional for `FLAC` and `WAV` audio files and required 411 # for all other audio formats. For details, see AudioEncoding. 412 "enableAutomaticPunctuation": True or False, # *Optional* If 'true', adds punctuation to recognition result hypotheses. 413 # This feature is only available in select languages. Setting this for 414 # requests in other languages has no effect at all. 415 # The default 'false' value does not add punctuation to result hypotheses. 416 # Note: This is currently offered as an experimental service, complimentary 417 # to all users. In the future this may be exclusively available as a 418 # premium feature. 419 "alternativeLanguageCodes": [ # *Optional* A list of up to 3 additional 420 # [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) language tags, 421 # listing possible alternative languages of the supplied audio. 422 # See [Language Support](/speech-to-text/docs/languages) 423 # for a list of the currently supported language codes. 424 # If alternative languages are listed, recognition result will contain 425 # recognition in the most likely language detected including the main 426 # language_code. The recognition result will include the language tag 427 # of the language detected in the audio. 428 # Note: This feature is only supported for Voice Command and Voice Search 429 # use cases and performance may vary for other use cases (e.g., phone call 430 # transcription). 431 "A String", 432 ], 433 "enableSeparateRecognitionPerChannel": True or False, # This needs to be set to `true` explicitly and `audio_channel_count` > 1 434 # to get each channel recognized separately. The recognition result will 435 # contain a `channel_tag` field to state which channel that result belongs 436 # to. If this is not true, we will only recognize the first channel. The 437 # request is billed cumulatively for all channels recognized: 438 # `audio_channel_count` multiplied by the length of the audio. 439 "enableWordTimeOffsets": True or False, # *Optional* If `true`, the top result includes a list of words and 440 # the start and end time offsets (timestamps) for those words. If 441 # `false`, no word-level time offset information is returned. The default is 442 # `false`. 443 "enableSpeakerDiarization": True or False, # *Optional* If 'true', enables speaker detection for each recognized word in 444 # the top alternative of the recognition result using a speaker_tag provided 445 # in the WordInfo. 446 # Note: Use diarization_config instead. This field will be DEPRECATED soon. 447 "maxAlternatives": 42, # *Optional* Maximum number of recognition hypotheses to be returned. 448 # Specifically, the maximum number of `SpeechRecognitionAlternative` messages 449 # within each `SpeechRecognitionResult`. 450 # The server may return fewer than `max_alternatives`. 451 # Valid values are `0`-`30`. A value of `0` or `1` will return a maximum of 452 # one. If omitted, will return a maximum of one. 453 "profanityFilter": True or False, # *Optional* If set to `true`, the server will attempt to filter out 454 # profanities, replacing all but the initial character in each filtered word 455 # with asterisks, e.g. "f***". If set to `false` or omitted, profanities 456 # won't be filtered out. 457 "useEnhanced": True or False, # *Optional* Set to true to use an enhanced model for speech recognition. 458 # If `use_enhanced` is set to true and the `model` field is not set, then 459 # an appropriate enhanced model is chosen if: 460 # 1. project is eligible for requesting enhanced models 461 # 2. an enhanced model exists for the audio 462 # 463 # If `use_enhanced` is true and an enhanced version of the specified model 464 # does not exist, then the speech is recognized using the standard version 465 # of the specified model. 466 # 467 # Enhanced speech models require that you opt-in to data logging using 468 # instructions in the 469 # [documentation](/speech-to-text/docs/enable-data-logging). If you set 470 # `use_enhanced` to true and you have not enabled audio logging, then you 471 # will receive an error. 472 "sampleRateHertz": 42, # Sample rate in Hertz of the audio data sent in all 473 # `RecognitionAudio` messages. Valid values are: 8000-48000. 474 # 16000 is optimal. For best results, set the sampling rate of the audio 475 # source to 16000 Hz. If that's not possible, use the native sample rate of 476 # the audio source (instead of re-sampling). 477 # This field is optional for FLAC and WAV audio files, but is 478 # required for all other audio formats. For details, see AudioEncoding. 479 "diarizationSpeakerCount": 42, # *Optional* 480 # If set, specifies the estimated number of speakers in the conversation. 481 # If not set, defaults to '2'. 482 # Ignored unless enable_speaker_diarization is set to true." 483 # Note: Use diarization_config instead. This field will be DEPRECATED soon. 484 "enableWordConfidence": True or False, # *Optional* If `true`, the top result includes a list of words and the 485 # confidence for those words. If `false`, no word-level confidence 486 # information is returned. The default is `false`. 487 "model": "A String", # *Optional* Which model to select for the given request. Select the model 488 # best suited to your domain to get best results. If a model is not 489 # explicitly specified, then we auto-select a model based on the parameters 490 # in the RecognitionConfig. 491 # <table> 492 # <tr> 493 # <td><b>Model</b></td> 494 # <td><b>Description</b></td> 495 # </tr> 496 # <tr> 497 # <td><code>command_and_search</code></td> 498 # <td>Best for short queries such as voice commands or voice search.</td> 499 # </tr> 500 # <tr> 501 # <td><code>phone_call</code></td> 502 # <td>Best for audio that originated from a phone call (typically 503 # recorded at an 8khz sampling rate).</td> 504 # </tr> 505 # <tr> 506 # <td><code>video</code></td> 507 # <td>Best for audio that originated from from video or includes multiple 508 # speakers. Ideally the audio is recorded at a 16khz or greater 509 # sampling rate. This is a premium model that costs more than the 510 # standard rate.</td> 511 # </tr> 512 # <tr> 513 # <td><code>default</code></td> 514 # <td>Best for audio that is not one of the specific audio models. 515 # For example, long-form audio. Ideally the audio is high-fidelity, 516 # recorded at a 16khz or greater sampling rate.</td> 517 # </tr> 518 # </table> 519 "diarizationConfig": { # *Optional* Config to enable speaker diarization and set additional 520 # parameters to make diarization better suited for your application. 521 # Note: When this is enabled, we send all the words from the beginning of the 522 # audio for the top alternative in every consecutive STREAMING responses. 523 # This is done in order to improve our speaker tags as our models learn to 524 # identify the speakers in the conversation over time. 525 # For non-streaming requests, the diarization results will be provided only 526 # in the top alternative of the FINAL SpeechRecognitionResult. 527 "minSpeakerCount": 42, # *Optional* Only used if diarization_speaker_count is not set. 528 # Minimum number of speakers in the conversation. This range gives you more 529 # flexibility by allowing the system to automatically determine the correct 530 # number of speakers. If not set, the default value is 2. 531 "enableSpeakerDiarization": True or False, # *Optional* If 'true', enables speaker detection for each recognized word in 532 # the top alternative of the recognition result using a speaker_tag provided 533 # in the WordInfo. 534 "maxSpeakerCount": 42, # *Optional* Only used if diarization_speaker_count is not set. 535 # Maximum number of speakers in the conversation. This range gives you more 536 # flexibility by allowing the system to automatically determine the correct 537 # number of speakers. If not set, the default value is 6. 538 }, 539 "speechContexts": [ # *Optional* array of SpeechContext. 540 # A means to provide context to assist the speech recognition. For more 541 # information, see [Phrase Hints](/speech-to-text/docs/basics#phrase-hints). 542 { # Provides "hints" to the speech recognizer to favor specific words and phrases 543 # in the results. 544 "phrases": [ # *Optional* A list of strings containing words and phrases "hints" so that 545 # the speech recognition is more likely to recognize them. This can be used 546 # to improve the accuracy for specific words and phrases, for example, if 547 # specific commands are typically spoken by the user. This can also be used 548 # to add additional words to the vocabulary of the recognizer. See 549 # [usage limits](/speech-to-text/quotas#content). 550 # 551 # List items can also be set to classes for groups of words that represent 552 # common concepts that occur in natural language. For example, rather than 553 # providing phrase hints for every month of the year, using the $MONTH class 554 # improves the likelihood of correctly transcribing audio that includes 555 # months. 556 "A String", 557 ], 558 "boost": 3.14, # Hint Boost. Positive value will increase the probability that a specific 559 # phrase will be recognized over other similar sounding phrases. The higher 560 # the boost, the higher the chance of false positive recognition as well. 561 # Negative boost values would correspond to anti-biasing. Anti-biasing is not 562 # enabled, so negative boost will simply be ignored. Though `boost` can 563 # accept a wide range of positive values, most use cases are best served with 564 # values between 0 and 20. We recommend using a binary search approach to 565 # finding the optimal value for your use case. 566 }, 567 ], 568 "metadata": { # Description of audio data to be recognized. # *Optional* Metadata regarding this request. 569 "recordingDeviceType": "A String", # The type of device the speech was recorded with. 570 "originalMediaType": "A String", # The original media the speech was recorded on. 571 "microphoneDistance": "A String", # The audio type that most closely describes the audio being recognized. 572 "obfuscatedId": "A String", # Obfuscated (privacy-protected) ID of the user, to identify number of 573 # unique users using the service. 574 "recordingDeviceName": "A String", # The device used to make the recording. Examples 'Nexus 5X' or 575 # 'Polycom SoundStation IP 6000' or 'POTS' or 'VoIP' or 576 # 'Cardioid Microphone'. 577 "industryNaicsCodeOfAudio": 42, # The industry vertical to which this speech recognition request most 578 # closely applies. This is most indicative of the topics contained 579 # in the audio. Use the 6-digit NAICS code to identify the industry 580 # vertical - see https://www.naics.com/search/. 581 "audioTopic": "A String", # Description of the content. Eg. "Recordings of federal supreme court 582 # hearings from 2012". 583 "originalMimeType": "A String", # Mime type of the original audio file. For example `audio/m4a`, 584 # `audio/x-alaw-basic`, `audio/mp3`, `audio/3gpp`. 585 # A list of possible audio mime types is maintained at 586 # http://www.iana.org/assignments/media-types/media-types.xhtml#audio 587 "interactionType": "A String", # The use case most closely describing the audio content to be recognized. 588 }, 589 }, 590 "name": "A String", # *Optional* The name of the model to use for recognition. 591 } 592 593 x__xgafv: string, V1 error format. 594 Allowed values 595 1 - v1 error format 596 2 - v2 error format 597 598Returns: 599 An object of the form: 600 601 { # The only message returned to the client by the `Recognize` method. It 602 # contains the result as zero or more sequential `SpeechRecognitionResult` 603 # messages. 604 "results": [ # Output only. Sequential list of transcription results corresponding to 605 # sequential portions of audio. 606 { # A speech recognition result corresponding to a portion of the audio. 607 "languageCode": "A String", # Output only. The 608 # [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) language tag of the 609 # language in this result. This language code was detected to have the most 610 # likelihood of being spoken in the audio. 611 "alternatives": [ # Output only. May contain one or more recognition hypotheses (up to the 612 # maximum specified in `max_alternatives`). 613 # These alternatives are ordered in terms of accuracy, with the top (first) 614 # alternative being the most probable, as ranked by the recognizer. 615 { # Alternative hypotheses (a.k.a. n-best list). 616 "confidence": 3.14, # Output only. The confidence estimate between 0.0 and 1.0. A higher number 617 # indicates an estimated greater likelihood that the recognized words are 618 # correct. This field is set only for the top alternative of a non-streaming 619 # result or, of a streaming result where `is_final=true`. 620 # This field is not guaranteed to be accurate and users should not rely on it 621 # to be always provided. 622 # The default of 0.0 is a sentinel value indicating `confidence` was not set. 623 "transcript": "A String", # Output only. Transcript text representing the words that the user spoke. 624 "words": [ # Output only. A list of word-specific information for each recognized word. 625 # Note: When `enable_speaker_diarization` is true, you will see all the words 626 # from the beginning of the audio. 627 { # Word-specific information for recognized words. 628 "confidence": 3.14, # Output only. The confidence estimate between 0.0 and 1.0. A higher number 629 # indicates an estimated greater likelihood that the recognized words are 630 # correct. This field is set only for the top alternative of a non-streaming 631 # result or, of a streaming result where `is_final=true`. 632 # This field is not guaranteed to be accurate and users should not rely on it 633 # to be always provided. 634 # The default of 0.0 is a sentinel value indicating `confidence` was not set. 635 "endTime": "A String", # Output only. Time offset relative to the beginning of the audio, 636 # and corresponding to the end of the spoken word. 637 # This field is only set if `enable_word_time_offsets=true` and only 638 # in the top hypothesis. 639 # This is an experimental feature and the accuracy of the time offset can 640 # vary. 641 "word": "A String", # Output only. The word corresponding to this set of information. 642 "startTime": "A String", # Output only. Time offset relative to the beginning of the audio, 643 # and corresponding to the start of the spoken word. 644 # This field is only set if `enable_word_time_offsets=true` and only 645 # in the top hypothesis. 646 # This is an experimental feature and the accuracy of the time offset can 647 # vary. 648 "speakerTag": 42, # Output only. A distinct integer value is assigned for every speaker within 649 # the audio. This field specifies which one of those speakers was detected to 650 # have spoken this word. Value ranges from '1' to diarization_speaker_count. 651 # speaker_tag is set if enable_speaker_diarization = 'true' and only in the 652 # top alternative. 653 }, 654 ], 655 }, 656 ], 657 "channelTag": 42, # For multi-channel audio, this is the channel number corresponding to the 658 # recognized result for the audio from that channel. 659 # For audio_channel_count = N, its output values can range from '1' to 'N'. 660 }, 661 ], 662 }</pre> 663</div> 664 665</body></html>