Supertone Inc. has released Supertonic 3, a compact 99-million-parameter text-to-speech model that runs entirely offline on hardware as modest as a Raspberry Pi and, according to the developers, outperforms cloud services from ElevenLabs, OpenAI and Google Gemini on real-world text. The model is distributed as an open-weight ONNX file and requires no GPU or internet connection, positioning it as a direct challenge to the subscription-based cloud TTS market.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.