Select a Model and Transcribe
A 3-step recipe with Deepgram. We can help you get access.
Recipe · 3 steps
Select a Model and Transcribe
with Deepgram
List models, inspect one, and transcribe audio using it.
- List the latest public speech-to-text and text-to-speech models available on the platform.
- Read the detailed metadata for the chosen model to confirm its languages, architecture, and version before transcribing.
- Submit the hosted audio file for transcription using the selected model with punctuation and smart formatting enabled.
What you'll need to have ready
- apiKey: Deepgram API key used to authenticate requests.
- modelId: The model identifier to fetch metadata for.
- modelName: The model name to pass to transcription (e.g. nova-3).
- audioUrl: Publicly accessible URL of the audio file to transcribe.
For your developer
Arazzo workflow deepgram:select-model-and-transcribe, 3 operations: listModels → getModel → transcribePreRecordedAudio. Specs on APIs.io.