Speechall OpenAPI reference
Speechall provides a unified API for speech-to-text models from multiple providers. Use this reference to explore request parameters, response formats, and schemas for the API. The interactive reference below loads the current OpenAPI specification and lets you inspect each operation in detail.
Authentication and model selection
Create an API key in the Speechall dashboard and send it as a Bearer token in the Authorization header. Keep the key private and avoid exposing it in client-side code. The model list endpoint describes available models; select a model identifier that supports the languages and features your application needs.
Transcribe audio
Send a POST request to https://api.speechall.com/v1/transcribe with a model identifier and the audio file in the request body.
You can choose an output format such as text, JSON, SRT, or VTT. Consult the operation below for supported parameters, response schemas, and
errors before integrating it into your application.
For a walkthrough, see the getting started guide and the speech-to-text guide. You can also download the OpenAPI specification for use with your own tools.
Interactive API reference
The reference below requires JavaScript. The OpenAPI specification is also available from the download link above.