Open-source customizable AI voice dictation built on Pipecat
Tambourine is an open source, fully customizable voice dictation system that lets you control STT/ASR, LLM formatting, and prompts for inserting clean text into any app. I have been building this on the side for a few weeks. What motivated it was wanting a customizable version of Wispr Flow where I could fully control the models, formatting, and behavior of the system, rather than relying on a black box. Tambourine is built directly on top of Pipecat and relies on its modular voice agent framework. The back end is a local Python server that uses Pipecat to stitch together STT and LLM…
In plain words
Tambourine is an open-source voice dictation system that converts speech to text and inserts it into any application. Built on Pipecat's modular framework, it runs as a local Python server and lets users control speech recognition models, language models, formatting, and prompts rather than relying on proprietary alternatives. It's designed for developers and power users who want customization and transparency in their voice AI setup.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Tambourine is an open source, fully customizable voice dictation system that lets you control STT/ASR, LLM formatting, and prompts for inserting clean text into any app. I have been building this on the side for a few weeks. What motivated it was wanting a customizable version of Wispr Flow where I could fully control the models, formatting, and behavior of the system, rather than relying on a black box. Tambourine is built directly on top of Pipecat and relies on its modular voice agent framework. The back end is a local Python server that uses Pipecat to stitch together STT and LLM models into a single pipeline. This modularity is what makes it easy to swap providers, experiment with different setups, and maintain fine-grained control over the voice AI. I shared an early version with friends and recently presented it at my local Claude Code meetup. The response was overwhelmingly positive, and I was encouraged to share it more widely. The desktop app is built with Tauri. The front end is written in TypeScript, while the Tauri layer uses Rust to handle low level system integration. This enables the registration of global hotkeys, management of audio devices, and reliable text input at the cursor on both Windows and macOS. At a high level, Tambourine gives you a universal voice interface across your OS. You press a global hotkey, speak, and formatted text is typed directly at your cursor. It works across emails, documents, chat apps, code editors, and terminals. Under the hood, audio is streamed from the TypeScript front end to the Python server via WebRTC. The server runs real-time transcription with a configurable STT provider, then passes the transcript through an LLM that removes filler words, adds punctuation, and applies custom formatting rules and a personal dictionary. STT and LLM providers, as well as prompts, can be switched without restarting the app. The project is still under active development. I am working through edge cases and refining the UX, and there will likely be breaking changes, but most core functionality already works well and has become part of my daily workflow. I would really appreciate feedback, especially from anyone interested in the future of voice as an interface.
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, December 2025
the whole month →- GP
Life & fun · Dec 2025 · dosaygo-studio.github.io
- JG
Hi everyone! My name's Luke and I made the original Jmail here alongside Riley Walz. We had a ton of friends collaborate on building out more of the app suite last night in lieue of DOJ's "Epstein files" release. Please AMA!
Life & fun · Dec 2025 · jmail.world
PlanEat AI▲733AI turns your health goals into a 7-day menu & grocery list
AI · Dec 2025 · planeatai.com
- 2G
Community, All the HN belong to you. This is an archive of hacker news that fits in your browser. When I made HN Made of Primes I realized I could probably do this offline sqlite/wasm thing with the whole GBs of archive. The whole dataset. So I tried it, and this is it. Have Hacker News on your device. Go to this repo (https://github.com/DOSAYGO-STUDIO/HackerBook): you can download it. Big Query -> ETL -> npx serve docs - that's it. 20 years of HN arguments and beauty, can be yours forever. So they'll never die. Ever. It's the unkillable static archive of HN and it's…
Dev tools · Dec 2025 · hackerbook.dosaygo.com
