Nari Labs
TTS 2 models no stock voices based in United States open weights only data from provider API synced 2026-11-18
US-based AI lab for open research that started with speech AI. It released Dia, a 1.6B-parameter English dialogue TTS model that generates two-speaker conversations with nonverbal sounds and supports voice cloning from an audio prompt, under Apache-2.0, followed by Dia2.
Text-to-speech 2
| Compare | Model | Price |
|---|---|---|
nari-labs/Dia-1.6BDia-1.6B |
– | |
nari-labs/Dia-1.6B-0626Dia-1.6B-0626 |
– |