· IDEAS · SIGNAL LOG
Concepts and thinking on voice, computing, and the tools we build. Some are RFCs, some are half-formed sketches, all are in flight.
A practical, honest roundup of the best dictation and speech-to-text apps for Mac in 2026 — on-device vs cloud, pricing models, and which one fits your workflow. Includes Superwhisper, Wispr Flow, MacWhisper, Aqua Voice, and more.
Looking for an Aiko alternative? Aiko transcribes audio files on-device with Whisper. Talkie does live dictation and turns captures into structured, agent-ready data.
An honest Apple Dictation alternative comparison. Both work on-device. Where Talkie differs: structured local storage, a programmable CLI, and dictating into AI agents like Claude Code.
Looking for an Aqua Voice alternative? Talkie is on-device voice dictation for Mac, iPhone, and Apple Watch — with a programmable CLI that pipes your voice into scripts and AI agents.
A DictaFlow alternative for people who want dictation that stays on-device and exposes captures as structured data they can pipe into scripts and AI agents.
Looking for a Dragon dictation alternative for Mac? Dragon left the consumer Mac market years ago. Talkie is an on-device, Mac-native option that also feeds your agents.
Looking for a MacWhisper alternative? Talkie offers on-device voice dictation for Mac, iPhone, and Apple Watch — with a CLI that turns your captures into structured data for scripts and AI agents.
Looking for an Otter.ai alternative that keeps your voice on-device? Talkie is local-first dictation and capture for Mac and iPhone, with a programmable path into scripts and agents.
Looking for a Superwhisper alternative? Talkie runs speech-to-text on-device, syncs across Mac, iPhone, and Apple Watch, and hands your voice to the command line and your AI agents. An honest comparison.
Looking for a Talon Voice alternative? Talkie is on-device Mac dictation that turns speech into structured captures you pipe straight into Claude Code and other agents.
An honest Wispr Flow alternative for Mac and iPhone: Talkie is on-device, local-first dictation with a CLI that turns your voice into structured data for scripts and AI agents.
How we redesigned the Talkie homepage by running six parallel subagents in git worktrees, comparing directions side by side, and picking a winner.
Apple's FoundationModels framework on macOS 26 lets Talkie run AI processing entirely on-device — no API key, no network call, no cost per request.
We tested Apple's Foundation Models against 24 real Talkie use cases. Here's what works, what needs a prompt, and what stays free forever.
A draft RFC for making Talkie programmable from disk with rules, tools, workflows, and automations.
Turning two weeks of voice memos into an Obsidian vault with four LLM prompts. Ten minutes, end to end.
A 0.6B model matches the 1.7B on accuracy, runs 2.4x faster, and fits in 350MB. At some point the task stops needing more parameters.
VLM vs text-only, where to draw the line between code and model, and why picking the right model size is harder than it sounds.
Fine-tuning a 1.5B model to reconstruct shell commands from voice. 97% accuracy, 3GB of RAM, under a second on a phone.
The Talkie CLI turns your voice memos and dictations into structured data that any script, pipeline, or AI agent can work with.
The same voice memo, transcribed three different ways. From Apple Speech on iOS to Parakeet v3 with post-processing — each layer improves readability while preserving the speaker's intent.
GUI apps are growing command lines. Here's why we built one for a voice app, and what it lets you do with your data after you've spoken it.
On where my best ideas actually happen, and why my phone keeps missing them.