Topic
Technology
AI systems, automation, tooling — what I build and what I learn building it.
55 posts · Page 1 of 3
Sub-topics
Contextual Sentence Creator: a 3B model that writes ASR training sentences
A Granite 4.2 3B fine-tune that writes natural sentences using a given term, for recording speech recognition training data. 95.7% of sentences pass the acceptance check; GGUF build included.
Fritter: a Parakeet fine-tune for English tech dictation
Fritter is my Parakeet TDT 0.6B v2 fine-tune for English dictation without Hebrew: tech terms, numbers and brand names. Tech WER falls from 7.53% to 4.01% on my voice. Browser demo and GGUF builds.
Hey Yossi: a custom wake word for Home Assistant
Hey Yossi is a custom wake word for Home Assistant, trained on real recordings, in two builds: openWakeWord for the server and Pi, microWakeWord for ESPHome, Voice PE and Android. 95% recall in testing.
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.
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.
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.
Agent Task Break Detector: knowing when to start a fresh session
A small ModernBERT model that checks every message in an AI agent session and says whether it is a new task a fresh session could handle. Runs locally on CPU; v0.1, trained on synthetic data.
A Claude Code transcription plugin that replaced five of my projects
Transcription Pipeline is a public Claude Code plugin that turns voice notes and calls into transcripts, specs, minutes and PDFs in eight stages. It replaced two plugins and three MCP servers, and was specified in a voice note it then transcribed.
Using AI to decide my vote in Israel's 2026 election
How I turned a 42-minute voice note into an AI vote decision for Israel's 2026 election: the method in diagrams, my result, and a step-by-step guide to doing it yourself with any AI chat app. No list has a real tenant-protection policy.
The Linux mic slider lies: Mic Monitor shows the real gain
On Linux, the microphone volume slider is not your mic's analog gain: mine was 60 dB out. Mic Monitor is a KDE Plasma tray app that shows the real gain in dB, clipping and noise floor, tuned for dictation and speech recognition.
Adding paragraph breaks to speech-to-text output with Gemma 3 270M
A Gemma 3 270M fine-tune that adds paragraph breaks to flat speech-to-text output and changes nothing else. Small enough for on-device use, with an ONNX export. Break F1 0.78 on 26 held-out memos.
Bolding the key lines of a dictated email with Gemma 3 270M
A preliminary Gemma 3 270M fine-tune that bolds the key lines of a dictated email (the asks, the purpose, the key facts) and changes nothing else. Small enough for on-device use. Word F1 0.62 on six held-out emails.
Proximate Pseudonymizer: a small model that swaps identifying details for close substitutes
A proof-of-concept Qwen3-0.6B fine-tune that rewrites text with names, places and dates swapped for close, consistent substitutes (Daniel to David, Jerusalem to Tel Aviv). Runs on CPU. Still leaks 9.3% of details, so review every output.
Boxty v1.0: English dictation without the Hebrew
Boxty is the no-Hebrew sibling of my Jachnun code-switching model: Parakeet fine-tuned on my voice for plain English and tech dictation. Tech WER drops from 2.51% (stock) to 1.35%, and general English holds.
Do LLMs push back on conspiracy believers? A small first look
A small, hand-graded first look at my conspiracy-endorsement eval: 7 prompts, 7 models, one sample each. In this run two uncensored models helped a believer and an invented theory tripped up two more. Not a verdict on any model.
My Weird Prompts passes 6,000 episodes
My Weird Prompts, my fully automated AI podcast, crossed 6,000 episodes on 15 October 2026: 5,991 in the archive, 2,688 hours of audio, and new translated editions in eight languages.
Voice typing that understands Hebrish
Speech-to-text mangles the Hebrew words English speakers in Israel use every day. Hebrish fixes that: free on-device apps for Android, Linux and Windows, the Jachnun and Sabich models, results, and how you can help.
Whisper Hebrish Small ACFT: a code-switching model for FUTO Keyboard
A phone-sized Whisper Hebrish for FUTO Keyboard on Android. It halves the stock English model's error rate on English-with-Hebrew dictation (11.8% to 5.8% WER) and imports as a single 264 MB file.
Whisper Hebrish v2 Released
Version 2 of Whisper Hebrish, my Whisper fine-tune for English with Hebrew words mixed in. Code-switching WER drops from 14.7% (stock turbo) to 6.7%, and there are whisper.cpp builds.
inventory-spec-extractor: a 34 MB model for reading product spec lines
A small token-classification model that pulls voltages, connectors, materials and 22 other attributes out of one line of a product spec. Runs on CPU in about 20 ms per line. F1 0.85 against a regex baseline's 0.58.
Can speech-to-text spell ‘makolet’? I tested seven models on Hebrish
English speakers in Israel talk Hebrish, and dictation software mangles the Hebrew words. I built an open benchmark to measure it: my fine-tune gets two-thirds of them right, off-the-shelf models a third to under half.
How much bitrate does an audio-multimodal LLM actually need?
An eval across 12 audio-capable models on OpenRouter, 5 MP3 bitrates, 4 dictation samples. The default upload quality in most dictation apps is 2–4× overprovisioned — and the more interesting finding had nothing to do with bitrate.
My Claude Code Projects Index & Plugin Marketplace
A look at two open source projects I've been building: a curated index of 200+ Claude Code project templates and a plugin marketplace with ready-to-install extensions.
Home Assistant Shabbat and Yom Tov Dashboard
A Home Assistant dashboard for Shabbat and Yom Tov observance featuring candle lighting times, Red Alert integration, weather, Hebrew date, and parsha display.
MakoletTime: Home Assistant Grocery List with NFC
A Home Assistant project for creating visual grocery shopping lists using button cards and NFC tags, designed for Israeli household shopping.