Inworld TTS
high-quality, affordable, and low-latency TTS. Realtime TTS-2 speaks about 500 languages and dialects, and both models return first audio in under 100 ms. Both clone or design voices, from $7 per million characters. High-quality voice APIs are usually either expensive, slow, or both.

반응 (1)
비슷한 프로덕트
설명이 비슷한 순
Nari Qwen3-TTS and Qwen3-ASR
narilabs.comHigh accuracy, low latency and cost
Sam TTS
samtts.comRecreates the classic Microsoft SAM voice in the browser
MARS5
github.comopen-source, insanely prosodic TTS model
I built a voice agent from scratch that averages ~400ms end-to-end latency (phone stop → first syllable).
Voice-Pro
github.comAI Voice Cloning