Skip to main content
Set speaker on every TTS request. For a first Coda request, try astra; use the live catalog endpoints for the complete, current lineup.

Starter voices to try

These speakers are confirmed to work across the listed models:

Browse the live catalog

Both endpoints are public and update as voices ship.

Languages

As of 19 May 2026, Rime supports the following languages:

Coda Voices (Flagship)

Coda is Rime’s flagship model. Its featured voices in the Rime dashboard are organized into four styles, each with female and male options.

Professional

Formal

Casual

Energetic

All voices above are confirmed to work with modelId: coda. The voice-details endpoint contains the complete catalog, including non-English voices.

Arcana v3 Voices

Choose Coda for new projects and existing Arcana traffic. It scores higher in Rime’s human evaluations for naturalness, prosody, and artifact-free output. See Start with Coda.
Arcana v3 is Rime’s previous flagship: a multilingual model with 94 expressive voices spanning age groups, regional accents, cultural backgrounds, and energy profiles. See the full voice list and voice details for available Arcana speakers.

The 94 Arcana Flagship Voices

Arcana v3 includes 94 flagship voices across English (US regional, international) plus the other supported Arcana languages. See /data/voices/voice_details.json for the full catalog with demographics.

Mist v3 Voices

Mist v3 supports the same popular speaker lineup as Mist v2 with significantly lower TTFA. See the full voice list for available Mist v3 speakers.

Mist v2 Voices

Mist v2 prioritizes clarity, pronunciation accuracy, and high-throughput production. For new projects, prefer Mist v3. See the full voice list for available Mist v2 speakers.