We turned EmbeddingGemma 2 into a speech-to-text model(eg2.haku.sh)
404 points by amrrs | 86 comments
Landing page for Embedding Gemma Transcribe, a free macOS app that adapts Google's EmbeddingGemma 2 into an on-device speech-to-text model.
#speech-to-text on Hacker News
11 stories tagged speech-to-text, newest first.
Show HN: BetterWispr – Free, open-source dictation for Mac(betterwispr.com)
183 points by kartik017 | 105 comments
Landing page for BetterWispr, a free, open-source macOS dictation app that cleans up speech into text locally, with per-app writing styles.
tiny-audio, nanoGPT for speech-to-text(github.com)
8 points by mazesmazes | 0 comments
GitHub repository for tiny-audio, a speech-to-text system trainable for about $25 that joins a frozen speech encoder to a Qwen LLM via a small trained projector.
Gemini-3.5-Transcribe(blog.google)
338 points by k9294 | 119 comments
Google blog post announcing Gemini 3.5 Transcribe, a speech-to-text model for precise real-time transcription.
Transcribe.cpp(workshop.cjpais.com)
696 points by sebjones | 144 comments
Project page announcing transcribe.cpp, a ggml-based speech transcription library supporting recent models, from the author of the Handy speech-to-text app.
Show HN: Ghost Pepper – Local hold-to-talk speech-to-text for macOS(github.com)
436 points by MattHart88 | 190 comments
GitHub repository for Ghost Pepper, an open-source macOS hold-to-talk speech-to-text app using local WhisperKit and LLM cleanup models.
Show HN: I built a sub-500ms latency voice agent from scratch(ntik.me)
531 points by nicktikhonov | 152 comments
Blog post by Nick Tikhonov describing how he built a voice agent with ~400ms end-to-end latency using streaming STT, LLM, and TTS with turn detection.
Voxtral Transcribe 2(mistral.ai)
961 points by meetpateltech | 232 comments
Mistral AI announcement page for Voxtral Transcribe 2, its speech-to-text transcription model.
Show HN: Whispering – Open-source, local-first dictation you can trust(github.com)
533 points by braden-w | 135 comments
GitHub directory for Whispering, an open-source, local-first speech-to-text dictation app within the Epicenter monorepo, built with Tauri.
OpenAI charges by the minute, so speed up your audio(george.mand.is)
689 points by georgemandis | 209 comments
A blog post by George Mandis explaining that speeding up audio 2-3x with ffmpeg before sending it to OpenAI transcription cuts cost and time with little quality loss.
Show HN: Countless.dev – A website to compare every AI model: LLMs, TTSs, STTs(countless.dev)
328 points by ahmetd | 70 comments
Countless.dev is a free, open-source website for comparing AI models such as LLMs, TTS, and STT, with price comparison, a pricing calculator, and versus views.