Skip to main content

Speech to Text: start task

POST 
/api/technology/speech-to-text-whisper-enhanced

Start an Enhanced Speech to Text Built on Whisper task for a media file.

Enhanced Speech to Text Built on Whisper features​

  • Multiple models can be deployed in the Speech Platform. You can select which one the task uses with the model query parameter.
  • Multi-channel audio files are supported.
  • Channel id is included in individual transcription segments.
  • Language for transcription of the whole media file can be forced with the language query parameter, otherwise it is detected automatically.
  • Language switching can be activated with the language_switching query parameter. With this feature, the technology automatically identifies the predominant language spoken within each thirty-second segment of the file and uses it for transcribing that particular segment.
  • If language is set, language_switching is ignored.
  • A language explicitly passed as the language argument must be licensed. Otherwise, an error is raised.
  • When the language is autodetected or language_switching is used, only licensed languages are considered as the source for the transcription. In case the actually detected language is not licensed, the closest licensed language is used instead.
  • Processing can be limited to a specific time segment in the media file with query parameters.

Multiple models​

Whisper Speech to Text supports multiple models that can differ in performance, accuracy, and supported languages. Models can even be adapted to improve performance on domain-specific data. The list of deployed models can be found in the response to the Speech Platform Info endpoint. Look for the technologies entry with technology set to enhanced_speech_to_text_built_on_whisper; its models field lists the deployed models.

Request​

Responses​

Speech to Text task was accepted. Follow the Location header to poll for the task state.

Response Headers
    X-Locationdeprecated

    Use the Location header instead. Will be removed in version 2027.1.0.

    Example: /api/technology/speech-to-text-whisper-enhanced/123e4567-e89b-12d3-a456-426614174000
    Location

    A URL the client should poll for task state and result.

    Example: /api/technology/speech-to-text-whisper-enhanced/123e4567-e89b-12d3-a456-426614174000