Gradium Open-Sources Phonon: 31M-Parameter On-Device TTS With 1.7% Word Error Rate
thisdudelikesAI · x · 2026-10-20
Voice AI company Gradium has open-sourced Phonon, a family of open-weight text-to-speech models (31M and 40M parameters) that run fully on-device, offline and privately, small enough to ship in any app.
Key benchmark (500 sentences other models get wrong — phone numbers, emails, postal codes):
- Phonon 31M (English): 1.70% WER
- Phonon 40M multilingual: 2.11%
- Phonon 40M: 2.22%
- For comparison: Kokoro 3.50%, Kitten TTS 7.40%, NeuTTS Nano 8.71%
Gradium also extended its seed funding to $100M with NVIDIA among new investors, and opened a Bay Area office to scale real-time voice AI.
Related event: Gradium Open-Sources Phonon: Sub-50M On-Device TTS Models(7 posts)→
More from Infra
- Stanford CS336 explains GPU kernels: data movement, not math, is what costs you — stanfordnlp · 2026-10-21
- DeepSeek open-sources DeepInfer, its production inference framework powering V4.1-Flash — zhuohan123 · 2026-10-21
- AI bond issuance hits $419B in 2026, up 90% from all of 2025 — SumitGup · 2026-10-21
- AI Obsession Is Making Enterprises Neglect Cloud FinOps, Expert Warns — DavidLinthicum · 2026-10-21
- DeepSeek bottleneck is compute, not demand: 1GW cluster could train a far stronger model, says leak account — teortaxesTex · 2026-10-21
- Neoclouds now gross ~$38 per GPU-hour at 155 trillion tokens/day per cluster — basedjensen · 2026-10-21