1
- // Copyright 2021 Google LLC
1
+ // Copyright 2022 Google LLC
2
2
//
3
3
// Licensed under the Apache License, Version 2.0 (the "License");
4
4
// you may not use this file except in compliance with the License.
@@ -383,6 +383,15 @@ message RecognitionConfig {
383
383
// <td><b>Description</b></td>
384
384
// </tr>
385
385
// <tr>
386
+ // <td><code>latest_long</code></td>
387
+ // <td>Best for long form content like media or conversation.</td>
388
+ // </tr>
389
+ // <tr>
390
+ // <td><code>latest_short</code></td>
391
+ // <td>Best for short form content like commands or single shot directed
392
+ // speech.</td>
393
+ // </tr>
394
+ // <tr>
386
395
// <td><code>command_and_search</code></td>
387
396
// <td>Best for short queries such as voice commands or voice search.</td>
388
397
// </tr>
@@ -404,6 +413,16 @@ message RecognitionConfig {
404
413
// For example, long-form audio. Ideally the audio is high-fidelity,
405
414
// recorded at a 16khz or greater sampling rate.</td>
406
415
// </tr>
416
+ // <tr>
417
+ // <td><code>medical_conversation</code></td>
418
+ // <td>Best for audio that originated from a conversation between a
419
+ // medical provider and patient.</td>
420
+ // </tr>
421
+ // <tr>
422
+ // <td><code>medical_dictation</code></td>
423
+ // <td>Best for audio that originated from dictation notes by a medical
424
+ // provider.</td>
425
+ // </tr>
407
426
// </table>
408
427
string model = 13 ;
409
428
@@ -444,6 +463,8 @@ message SpeakerDiarizationConfig {
444
463
445
464
// Description of audio data to be recognized.
446
465
message RecognitionMetadata {
466
+ option deprecated = true ;
467
+
447
468
// Use case categories that the audio recognition request can be described
448
469
// by.
449
470
enum InteractionType {
@@ -819,6 +840,9 @@ message SpeechRecognitionResult {
819
840
// Alternative hypotheses (a.k.a. n-best list).
820
841
message SpeechRecognitionAlternative {
821
842
// Transcript text representing the words that the user spoke.
843
+ // In languages that use spaces to separate words, the transcript might have a
844
+ // leading space if it isn't the first result. You can concatenate each result
845
+ // to obtain the full transcript without using a separator.
822
846
string transcript = 1 ;
823
847
824
848
// The confidence estimate between 0.0 and 1.0. A higher number
0 commit comments