capabilities.work

Transcribe, Analyze, and Synthesize

A 3-step recipe with Deepgram. We can help you get access.

Recipe · 3 steps

Transcribe, Analyze, and Synthesize

with Deepgram

Transcribe audio, analyze the transcript text, and synthesize speech.

  1. Transcribe the hosted audio file with punctuation and smart formatting so the transcript is ready for text intelligence.
  2. Run text intelligence over the transcript to produce a summary along with sentiment, topics, and intents.
  3. Convert the generated summary text back into natural-sounding speech audio using the selected Aura voice.
What you'll need to have ready
  • apiKey: Deepgram API key used to authenticate all requests.
  • audioUrl: Publicly accessible URL of the audio file to transcribe.
  • sttModel: Speech-to-text model to use for transcription.
  • ttsModel: Text-to-speech Aura voice to synthesize the response with.
For your developer

Arazzo workflow deepgram:transcribe-analyze-synthesize, 3 operations: transcribePreRecordedAudio → analyzeText → synthesizeSpeech. Specs on APIs.io.

More recipes with Deepgram

Audit Project Usage and BillingBalance-Gated TranscriptionDiarized Transcript IntelligenceInvite and Confirm a Project MemberProvision a Project API KeyRotate a Project API KeySelect a Model and TranscribeTranscribe Audio and Track UsageUpdate a Member's Scopes