Audio8-ASR-Infinite: streaming ASR for unbounded audio
Real-time streaming speech-recognition model built for infinite-length audio, English and Chinese.
Why it matters
Apache-2.0 streaming ASR that handles unbounded recordings without chunking - rare among open ASR releases.
What you could build with it
A media monitoring startup can transcribe long podcasts, livestreams and earnings calls continuously, in English and Chinese, and alert clients when their brand is mentioned.
Does it hold up?
Too early to judge. Only the publisher's own repo and one independent review exist; there are no third-party integrations or builder demos, and mainstream inference engines (llama.cpp, vLLM, MLX) don't register the model yet.
Built with Audio8-ASR-Infinite
- Local Model Watch: Edge0 Releases Audio8-ASR-Infinitelocalmodelwatch · Independent review with a hardware verdict (12GB+ VRAM, 7.6GB BF16 weights), benchmark comparison vs Voxtral-Mini realtime and NVIDIA Nemotron streaming ASR (best average), and a note that llama.cpp, vLLM, and MLX registries don't cover it yet.
- Edge0-AI/Audio8-ASR-Infinitegithub · 73 ★ · Publisher's official repo (verified via Edge0-AI org listing): Docker/vLLM deployment and transformers inference with trust_remote_code for the 4.1B streaming ASR model, with sample outputs on the HF model card.
First spotted on huggingface: source.
More AI models
ElevenLabs Eleven v4 + v4 Turbo: new emotive TTS architecture with a 100ms real-time variantElevenLabs released Eleven v4 and Eleven v4 Turbo on September 28: a new text-to-speech architecture with inline…models · JEV 0.72VoiceStudio: open-source, fully-local ElevenLabs alternative with voice cloning, dubbing and dictation in 646 languagesVoiceStudio is an open-source, fully-local voice AI studio: voice cloning from a 3-second sample, voice design, video…models · JEV 0.71DeepSeek V4-Flash official API: public beta with upgraded agent capabilities, Responses API and Codex supportDeepSeek's official V4-Flash is now a public beta API with massively upgraded agent capabilities - benchmark scores…models · JEV 0.69Mistral Forge: enterprise platform for training and continuously improving proprietary modelsMistral AI launched Forge (Sep 28) — a full-lifecycle model training platform (pre-training, SFT, DPO/ODPO, RL…models · JEV 0.69kev: open, trainable Jev-like family of small decision models on Qwen3.5/3.8A family of small decision models (0.8B to 27B) built on Qwen3.5/3.8 that reproduces Jev's prefill-only typed-decision…models · JEV 0.68GPT-Synopsys: OpenAI and Synopsys multi-year deal to build an AI model that operates EDA toolsOpenAI and Synopsys signed a multi-year agreement on Sep 30 to jointly develop GPT-Synopsys, a specialized model that…models · JEV 0.66
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.