inworld's 64M voice isolator running fully in the browser on WebGPU, cleaning a 20 second clip in about 6 seconds
Summary
A developer added Inworld's 64M-parameter voice isolator v1 to their caption app, running fully in-browser on a handwritten WebGPU engine with fp16 weights, cleaning a 20-second clip in about 6 seconds with no uploads. The demo is live at cuecaps.com/studio, with noise removal free and export behind a paid plan.
Similar Articles
inworld/voice-isolator
Inworld released Voice Isolator v1 under Apache-2.0: two 48 kHz voice isolation models (offline and realtime with 20ms look-ahead) for cleaning noisy speech recordings, live streams, and speech datasets, with a Python package and CLI available on Hugging Face and GitHub.
Reverb-TTS: SOTA open-source zero-shot voice cloning running in the browser on WebGPU, 17x realtime on M4 Pro
Reverb-TTS is an open-source zero-shot voice cloning TTS model that runs fully locally in the browser on WebGPU via ruNNtime, achieving about 17x realtime on an M4 Pro with no server round-trips.
Floor for local meeting summarization on a 6GB GPU: qwen3.5:0.8b works at 57s, Granite 4 350M hallucinates
The author introduces VoiceFlow, an open-source local dictation and meeting transcription tool, and benchmarks small LLMs (qwen3.5:0.8b and Granite 4 350M) for meeting summarization on a 6GB GPU, finding the 0.8B Qwen viable while sub-500M models hallucinate. They also ask the community for long-context summarization solutions on low VRAM.
whisper base and an 11M param BERT i distilled from a 2B model, both running in the browser to caption videos
The creator built CueCaps Studio, a browser-based video captioning tool running Whisper base via transformers.js with WebGPU/WASM fallback, plus an 11M-param BERT tagger distilled from LFM3-Omni-2B that handles emphasis, line breaks, and punctuation.
parakeet.wgsl – Fast, accurate ASR in the browser, via raw WebGPU & SIMD WASM
parakeet.wgsl enables fast, accurate NVIDIA Parakeet TDT 0.6B V2 speech transcription entirely in the browser using raw WebGPU compute shaders and SIMD WASM, with a live demo and open-source library.