Owen has independently developed and funded Inflect-Micro-v2 and Inflect-Nano-v2, two new text-to-waveform speech synthesis models. These models are designed for local, fixed-voice English TTS, prioritizing either quality with under 10 million parameters (Micro) or a smaller footprint with under 4 million parameters (Nano). Both models demonstrate competitive performance in human preference studies and intelligibility tests, while also offering fast CPU inference. AI
IMPACT These compact TTS models could enable more on-device and real-time speech synthesis applications.
RANK_REASON Release of new, compact, open-source TTS models with detailed evaluation metrics.
Read on Hugging Face Trending Models →
- Inflect-Micro-v2
- KittenTTS Nano
- Nemotron 3.5
- Owen
- owensong/Inflect-Micro-v2
- Qwen3-ASR
- Supertonic 3
- Whisper large-v3
- Hugging Face
- owensong/Inflect-Nano-v2
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
