Synchronous recognition API
Written by
Updated at February 10, 2025
With the synchronous recognition API, you can transcribe prepared audio files with the following characteristics:
- Maximum file size: 1 MB
- Maximum duration: 30 seconds
- Maximum number of audio channels: 1
The synchronous recognition service is located at stt.api.cloud.yandex.net/speech/v1/stt:recognize
Query parameters
| Parameter | Description |
|---|---|
| lang | string Recognition language. See the model description for acceptable values. The default value is ru-RU, Russian. |
| topic | string Language model to use for recognition. The more accurate your choice of the model, the better the recognition result. You can specify only one model per request. The acceptable values depend on the language you select. The default value is general. |
| profanityFilter | boolean This parameter controls the profanity filter in recognized speech. Acceptable values:
|
| rawResults | boolean Flag for how to write numbers: true for words, false (default) for figures. |
| format | string Submitted audio format. Acceptable values:
|
| sampleRateHertz | string Submitted audio sampling frequency. Applies if format equals lpcm. Acceptable values:
|
| folderId | string ID of the folder you have access to. Required for authorization with a user account (see Authentication with the SpeechKit API resource). Do not use this field if you make a request on behalf of a service account. The maximum string length is 50 characters. |
Parameters in the request body
The request body has to contain the binary content of an audio file.
Response
The recognized text is returned in the result field of the response.
{
"result": <recognized_text>
}
For more information about the response format and codes, see Response status codes.