Transcribe Audio and Track Usage
A 3-step recipe with Deepgram. We can help you get access.
Recipe · 3 steps
Transcribe Audio and Track Usage
with Deepgram
Transcribe an audio URL then reconcile it against project usage and requests.
- Submit the hosted audio URL for transcription with punctuation and smart formatting enabled, capturing the request identifier from the metadata.
- Retrieve the aggregated usage summary for the project over the supplied time window to confirm the audio hours were recorded.
- List the project's recent API requests for the same window so the transcription request_id can be located in the audit log.
What you'll need to have ready
- apiKey: Deepgram API key used to authenticate requests.
- projectId: The Deepgram project the usage and request logs belong to.
- audioUrl: Publicly accessible URL of the audio file to transcribe.
- model: Speech-to-text model to use for transcription (e.g. nova-3).
- start: Start of the usage/request window in ISO 8601 format.
- end: End of the usage/request window in ISO 8601 format.
For your developer
Arazzo workflow deepgram:transcribe-and-track-usage, 3 operations: transcribePreRecordedAudio → getProjectUsage → listProjectRequests. Specs on APIs.io.