Summary

Options including keyterms, paragraphs, intents, sentiment, and replacement are now sent to Deepgram instead of being silently ignored, while diarisation is now opt-in rather than enabled by default. The provider also composes voice identifiers, passes speech speed, exposes response metadata, and parses Deepgram’s structured error responses.

Why it matters

The update prevents silent request mismatches in production voice applications and makes billing, diagnostics, and speaker-diarisation behaviour more explicit.

Sources