By “SOTA” tts I think you mean LLM based TTS? With sound and language tokens tra... | Hacker News

Hacker News new | past | comments | ask | show | jobs | submit

login

jameszhao00 on July 16, 2023 | parent | context | favorite | on: SoundStorm: Efficient Parallel Audio Generation

By “SOTA” tts I think you mean LLM based TTS? With sound and language tokens trained GPT style?

Without going into too much details, imo they’re not really usable right now for TTS use cases.

og_kalu on July 18, 2023 [–]

Not necessarily LLM style. The above isn't for instance.

also Google Studio Voices is excellent. Definitely better than Microsoft's best, albeit very limited voices.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact