9:41
/ Z Nordhaven Labs
Signed-in account portrait — photo by Ben Khatry on Pexels
Voice tools Library Line NEW Studio Platform

Changelog

Support
2025-09-02

Model updates

  • New model: aster-3-preview
  • Controllability: aster-3-preview ships with finer controls for pace and warmth, so a single voice can carry a much wider range of takes. Full parameter list lives in the docs.
2025-04-14

Model updates and feature changes

  • Embeddings support: voice embeddings are retired in our latest models. Dropping them measurably improved generation quality and run-to-run stability.
  • Voice controls: Pace and Emotion are paused while we harden the newest model. If you still need them, call the API with aster-2-2025-03-07 and follow the migration note in the reference.
  • Voice cloning: current models support the similarity method only, which returns the closest match to the source take.
  • Voice designer: discontinued — it depended entirely on embeddings.
2025-02-20

Latency and regions

  • Streaming: first-audio latency is down to 214 ms at p50 for aster-2 on warm connections.
  • Regions: the eu-central endpoint has left preview and now carries the same uptime terms as us-east.