Comment by oidar

Any plans to offer speech to speech models which keep prosody, intonation, and timing intact? ElevenLabs is getting expensive for this.