Welcome
The Gladia Enterprise API turns speech into text while it is spoken. You open a session, stream audio over a WebSocket, and receive transcripts in a fraction of a second: partial ones while a sentence is spoken, a final one when it ends. On the same connection you can get translations, named entities, redacted personal data, sentiment and end-of-turn events.
How it works
Section titled “How it works”- Create a session with
POST /v2/live: its audio format, model, languages and options. The answer has a WebSocket URL. - Stream the audio to that URL, in small chunks, as it is captured.
- Read the messages:
transcriptmessages, partial then final, and the messages of the features you turned on. - Stop with
stop_recording: the last transcripts arrive, then the whole session’s result.
The session’s transcript stays available afterwards (GET /v2/live/{id}), in your
history, and in Gladia Workspace.
What you can build
Section titled “What you can build”- Voice agents: low-latency transcripts and end-of-turn detection, so the agent answers when the caller has finished, not at every pause.
- Call centers: one speaker per channel, PII redaction, sentiment, named entities.
- Meetings and live captions: speaker diarization, translation into 38 languages, custom vocabulary for names and jargon.
For LLMs and agents
Section titled “For LLMs and agents”Every page has a Markdown twin (add .md to its address), /llms.txt indexes them,
/llms-full.txt has the whole documentation in one file, and /openapi.json is the
OpenAPI document of the API.
- Quickstart: your first transcript.
- Playground: try it with your microphone.
- Models: pick the right one.