AI Productivity

Voice Mode for Agent Input

Say your prompt and the words land in Agent Input at the caret, ready to edit or send. Speech is turned into text on your computer by OpenAI Whisper, either built into 1DevTool or in 1AudioTool, and your audio is never uploaded.

Voice Mode for Agent Input

More Screenshots

Click any image to zoom in

The voice mode chooser comparing Built-in and 1AudioTool, with pros, cons and the Whisper model sizes
Settings → Agents → Voice with the engine, Built-in models, language, live transcript, microphone and activation options

Why You Need This

How this feature boosts your daily productivity

The prompts that get good work from a coding agent are long and specific, and typing them is the slow part, so they get shorter than they should be. Dictation apps paste into whatever has focus and often send your audio to a server. Voice mode puts speech-to-text inside Agent Input: it knows where your caret is, shows the words as they are recognised, and lets you undo a take in one click.

How It Works

Get started in 4 simple steps

1
1

Click the mic or press ⌘⇧Space

The microphone sits next to the attach button in Agent Input; Ctrl+Shift+Space on Windows and Linux. The first click asks How should voice mode listen?

2
2

Pick Built-in or 1AudioTool

Built-in runs Whisper inside 1DevTool after a one-time download of Tiny, Base or Small. 1AudioTool runs it in its own app and pastes into Agent Input.

3
3

Talk, pause, watch it arrive

The listening strip shows a level meter and a clock. Live shows the words as they are recognised; Pause closes the mic until Continue.

4
4

Insert, Send or Cancel

Enter inserts at the caret, ⌘⏎ inserts and sends, Esc drops the take. Inserted · Undo takes the words back out.

Without 1DevTool

The manual way vs. the 1DevTool way

The Hard Way

# type the whole prompt by hand
# or: open a dictation app, speak,
# click back into Agent Input,
# paste, fix where it landed

The 1DevTool Way

Press ⌘⇧Space, say it, press Enter: the words land at the caret, ready to edit

Key Capabilities

Everything included with Voice Mode for Agent Input

Dictate a prompt into Agent Input at the caret
Choose Built-in Whisper or 1AudioTool, with pros and cons side by side
Download or delete Tiny, Base and Small models
Watch a live transcript while you speak
Pause and continue a take without losing it
Finish, insert and send in one step with ⌘⏎
Undo an inserted take in one click
Hold the shortcut to talk, or click to start and stop

Why It Matters

The productivity impact of having ai productivity built into your workspace

Long Prompts at Speaking Speed

You say the file, the constraint and the test you want in a few seconds, so the detailed prompt stops being the expensive one.

Your Audio Stays Here

Whisper runs on your computer, built in or in 1AudioTool, so nothing you say is sent to a server.

You Read It Before the Agent Does

Insert lands at the caret without sending, highlighted and one click from Undo; only Send or ⌘⏎ sends.

Ready to boost your workflow?

Download 1DevTool for free and experience Voice Mode for Agent Input along with 347 other features — all in one app.