Skip to content

Transcripts

Who said what, after the call

Each participant's microphone is transcribed on its own, so every line has a speaker. Turn it on for a project, or request it for one session after the call ends.

Transcripts

  • One line, one speaker

    Speech-to-text runs on each participant's microphone separately. The speaker on a line is that person's user id, the one you set on their connection token.

  • On for a project, or one session

    Turn it on in the dashboard, or with PUT /v1/projects/:id/transcription (enabled and language). Or request one session that has ended with POST /v1/sessions/:id/transcript. Asking again returns the transcript already queued, running or finished.

  • Per speaker, or one language

    auto detects each speaker's language. multi is for someone who switches languages mid-sentence. A BCP-47 tag such as en or pt-BR transcribes everyone in that language.

  • JSON, subtitles, or captions

    GET /v1/sessions/:id/transcript returns the cues. Add ?format=vtt, srt or txt, and ?lang= for one language. Replay in @synento/player shows the same cues as captions.

  • API, SDK, CLI and MCP

    Node: requestTranscript(). CLI: synento transcript --request. MCP: request_transcript and get_transcript. Subscribe to transcript.ready instead of polling.

  • Billed on speech, not silence

    Silence and muted stretches are removed before anything is sent, and only that speech is billed, summed over participants. Both plans include 60 minutes a month. Past that, Free stops and pay-as-you-go is $0.006 per transcribed minute.

  • Deleted with the session

    Audio is sent to Deepgram only for a transcript you turned on or asked for, with model training opted out. The transcript is stored with the recording, deleted with the session, and redacting a participant removes their lines.

terminal
# Turn transcripts on for every session in a project.
curl -X PUT https://api.synento.com/v1/projects/$PROJECT_ID/transcription \
  -H "Authorization: Bearer $SYNENTO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"enabled": true, "language": "auto"}'

# Or request one session after it has ended. Asking again is safe.
curl -X POST https://api.synento.com/v1/sessions/$SESSION_ID/transcript \
  -H "Authorization: Bearer $SYNENTO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"language": "auto"}'

# Read it. format is vtt, srt or txt. lang keeps one language.
curl -s "https://api.synento.com/v1/sessions/$SESSION_ID/transcript?format=vtt" \
  -H "Authorization: Bearer $SYNENTO_API_KEY"

Record the calls your product runs on.

Sign up for a free account. Replay, MP4 export and post-call transcripts are included, and a coding agent can set the integration up from llms.txt.