Your hands are busy — you're driving, jogging, or cooking dinner. A task flashes through your mind, and a minute later it's gone. Sound familiar? Voice input in AI Planner solves exactly this problem: tap the microphone, say the task out loud — and within seconds it appears in your Eisenhower matrix with a title, quadrant, and tags.
Why voice input in a planner
Text input is great when you're at a computer. Screenshots save you when a task arrives in a messenger. But there are moments when neither a keyboard nor a screen is available. Voice is the fastest way to capture a thought: the average time to dictate a task is 10 seconds.
Typical scenarios:
- You're driving and remember you need to call a client
- A project idea comes to you during a walk
- Between meetings you need to quickly capture a conversation outcome
- Your hands are dirty after gardening, but a task is urgent
How it works
Under the hood, three steps happen automatically:
- Voice recording. You tap the microphone button in the planner. The browser records audio via the standard Web Audio API.
- Speech recognition. The audio file is sent to the server, where the Whisper model processes it — one of the best open-source transcription models. It supports Russian, English, and dozens of other languages.
- AI task parsing. The transcribed text is passed to a language model that extracts the task title, determines the Eisenhower matrix quadrant, assigns tags, and sets a deadline if a date is mentioned.
The result is a ready-made task card that you can edit right away or leave as is.
Voice, text, or screenshot — which to choose
AI Planner supports three ways to create tasks. Each excels in its own situation:
- Text — when you're at a computer and can formulate the task in writing. Maximum accuracy.
- Screenshot — when the task arrived in a messenger, email, or document. Just take a screenshot.
- Voice — when your hands are busy or you don't have time to type. The fastest method.
All three methods produce the same result: a structured card with a quadrant, tags, and title.
Practical tips
To get the most accurate recognition, follow a few simple rules:
- Speak in a quiet place. Background noise reduces transcription accuracy. If you're outside, bring the phone closer.
- Keep it short. "Call Marina about the budget by Friday" is better than a long monologue with digressions.
- Mention context. If you say "urgent" or "by Friday," the AI will pick the right quadrant and deadline.
- Review the result. AI determines the quadrant automatically, but you can always adjust it manually.
Voice and the Eisenhower matrix
One of the biggest barriers to using the Eisenhower matrix is the entry friction. You need to open the app, create a task, choose a quadrant, write a description. Voice input lowers this barrier to a minimum: you simply say what needs to be done, and the AI determines the priority.
This is especially important for quadrant Q2 ("important, not urgent"). Tasks in this quadrant are often lost because nobody writes them down — after all, they're not on fire. Voice input makes capturing them so easy that even a fleeting thought can be saved in 10 seconds.
Try it yourself
Voice input is available for free in AI Planner. Open the planner, tap the microphone icon, and dictate your first task. Within seconds it will appear in your Eisenhower matrix — with a title, quadrant, and tags.