Gradium open-sources Phonon, sub-50M TTS models under Apache 2.0
mishig25 · x · 2026-10-21
Speech startup Gradium has open-sourced Phonon, a suite of on-device TTS models all under 50M parameters, claiming a new standard for naturalness and accuracy across languages with broad inference support.
Phonon-31m is available under Apache 2.0. Former Hugging Face robotics lead Remi Cadene called it 'the best open-source TTS model.'
Related event: Gradium Open-Sources Phonon: Sub-50M On-Device TTS Models(8 posts)→
More from Multimodal
- AI film goes from idea to IMAX premiere and greenlit series in 14 days — lmoroney · 2026-10-21
- Anam blind benchmark: 68% positive lip-sync ratings vs ~40% for rivals on 960 conversations — SucceededMind · 2026-10-21
- AI art demo: combining Midjourney with Seedance for striking results — gizakdag · 2026-10-21
- Higgsfield launches Genjutsu Lite to insert your photo into trending videos at 3.5x lower cost — tisch_eins · 2026-10-21
- Uncensored image-to-image Space trends on Hugging Face with Gradio and MCP support — fishpasters · 2026-10-21
- Orchestrating open-source video models with Opus 5.5 cuts costs ~80x vs Seedance — mickeyxfriedman · 2026-10-21