Tag

asr-post-processing

3 posts · all tags

S1-ASR-Cleanup: cleaning up dictation with a fine-tuned S1-mini

S1-ASR-Cleanup: cleaning up dictation with a fine-tuned S1-mini

A fine-tune of S1-mini by Superwhisper that cleans speech-to-text dictation: paragraphs, dates and money as digits, contractions and fillers handled. 0.6B, with GGUF builds for llama.cpp and phones.

ai-models hugging-face fine-tuning +3
Proofed: could a speech model proofread its own transcript?

Proofed: could a speech model proofread its own transcript?

An idea in development: fine-tune Parakeet on cleanup alone, so dictation comes out without fillers, with digits and paragraphs, in one pass and with no LLM afterwards. Research done, data being recorded.

ideas proof-of-concept speech-recognition +3
Update: a 32 MB model for ASR paragraph breaks
Update

Update: a 32 MB model for ASR paragraph breaks

MiniLM-L12-ASR-Paragraph-Breaks replaces my Gemma 3 270M paragraph model: break F1 0.85 against 0.78, 32 MB instead of 437 MB, and it cannot change a word of the transcript.

ai-models hugging-face dictation +3