Dictum SDK

Voice capture for apps and agents.

Use the Dictum widget or headless SDK to add voice capture, transcription states, and transcript events without exposing raw provider keys or endpoint plumbing.

Read the SDK docs See pricing

Built for real voice workflows.

  • Widget UI or headless client
  • Shared listening and recording states
  • Server-side model and key profile

For products

Build voice capture into customer-facing apps with predictable SDK states and transcript events.

For agents

Expose speech-to-text as a client capability without making internal transport the product contract.

For teams

Centralize transcription quality, privacy, and model routing in one server-side profile.