Skip to content
View kdelmotte's full-sized avatar

Block or report kdelmotte

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
kdelmotte/README.md

Hey, I'm Kevin.

I'm a product manager who builds things. These repos are both side projects and tools I use every single day to work faster.

Mumble — Hold Fn, talk, release. Mumble transcribes your speech via Groq Whisper, detects the app you're in (email? Slack? VS Code?), and formats the output to match the context. Menu bar app, zero dependencies, signed and notarized.

TypoFixr — Cmd+Shift+D (or the shortcut of your choice) in any text field on macOS. Fixes typos and grammar while preserving your voice, no rewrites, just corrections. Parallel sentence chunking under the hood, to make it lightning fast.

Why I build: I think the best PMs understand the medium they're working in. Both apps are wired into Groq's inference APIs, Whisper large-v3 for transcription, reasoning models with tuned token budgets and temperature-zero determinism for text correction. I care about the details that make AI feel invisible: context-aware prompt routing, adaptive reasoning effort based on input complexity, and hallucination filtering on silent audio. These started as scratching my own itch and turned into real, shipped software, code-signed, notarized, and distributed via DMG.

I'm always looking for the next piece of daily friction to automate. If you try either app, I'd love to hear what you think, open an issue or reach out!

Pinned Loading

  1. Mumble Mumble Public

    A simple, fast and free speech to text app running on OpenAI's Whisper Large v3

    Swift