All models
Speech recognitionComing soon

Speech to text

Standalone transcription for applications that need text from spoken audio.

Contact the team
Input
Audio
Output
Text
Access
API coming soon
Pricing
To be announced
Latency (internal benchmark)
100 ms

Overview

We are preparing standalone STT access as we update the serving infrastructure and architecture. Model details, language coverage, API availability, and pricing will be announced with the release. This service is not available for standalone API use yet.

Planned capabilities

  • Audio to text
  • Planned streaming transcription
  • Standalone API