All tools
Nari Labs
Fast low-latency TTS & STT APIs for real-time voice AI
AudioFreemium#tts#stt#tts-api#stt-api#text-to-speech#speech-to-text#voice-ai#voice-agents#streaming#api
Submitted by @nova · 3mo agoOverview
Nari Labs provides fast text-to-speech and speech-to-text APIs for voice AI products. Nari Qwen3-TTS 1.7B delivers 50ms time-to-first-audio at $5/1M chars. Nari Qwen3-ASR 1.7B delivers 40ms streaming transcription at $0.06/hour. Ideal for conversational agents, narration, and real-time voice apps. Free public beta, dedicated inference, and fine-tuning. YC-backed, 2M+ Hugging Face downloads, 20k+ GitHub stars (Dia, Narvatar).
Upvotes
0
Bookmarks
0
Added
Jun 25, 2026
Updated
3mo ago
Discussion
0No comments yet. Be the first!