S1-mini
PulseAugur coverage of S1-mini — every cluster mentioning S1-mini across labs, papers, and developer communities, ranked by signal.
- 2026-08-19 product_launch Superwhisper released the S1-mini, a local browser-based LLM for transcript refinement. source
- 2026-08-19 product_launch Superwhisper released the S1-mini, a 0.6B parameter text normalization model. source
- 2026-08-19 product_launch Superwhisper released S1-mini, an open-weights text normalization model. source
- 2026-08-19 product_launch Cohere launched S1-mini, its first open-weights language model designed for on-device processing. source
-
RAG extensions amplify ASR errors, new research finds · 6 sources tracked
Research indicates that advanced retrieval-augmented generation (RAG) techniques, while improving overall performance, can amplify errors originating from Automatic Speech Recognition (ASR) systems. Specifically, multi-…
-
Superwhisper releases S1-mini, an open-weights text normalizer for ASR transcripts
Superwhisper has released S1-mini, an open-weights text normalization model designed to clean up raw Automatic Speech Recognition (ASR) transcripts. This 0.6B parameter model, fine-tuned from Qwen/Qwen3-0.6B, removes fi…
-
Cohere releases S1-mini, its first open-weights local language model
Cohere has released S1-mini, its first open-weights language model. This 0.6 billion parameter model is designed to process transcripts entirely on a user's device, emphasizing local hosting and local creation. The mode…
-
Superwhisper releases S1-mini for cleaning speech-to-text transcripts
Superwhisper has released S1-mini, a 0.6 billion parameter language model designed to clean up raw speech-to-text transcripts. The model removes filler words, corrects self-corrections, adds punctuation and capitalizati…