English, Russian, and Ukrainian
Interface and recognition for English, Russian, and Ukrainian. Auto mode can work with other Whisper languages.
Voice input for developers
No copying. No pasting. Straight into the active app.
What VoiceToPrompt can do
Voice input for prompts, tasks, and notes: local or cloud, with project files, term dictionaries, and formatted output.
Dictation
Local or cloud
Local recognition on Apple Silicon through WhisperKit, or cloud recognition through your OpenAI API key.
Offline mode
Local recognition works without an internet connection.
Accuracy
Term dictionaries
Corrects technical terms and tool names after recognition.
Retry recognition
Run recognition again from a saved recording with another engine or model.
AI Fix-It
Fixes recognition mistakes without rewriting the meaning of your text.
Workflow
Clipboard input
Insert clipboard content into the final prompt.
Files and projects
Connect a project and mention files while dictating — VoiceToPrompt adds the file name or path to the final prompt.
Markdown and Rich Text support
Insert marked-up text while dictating a prompt and get formatted output as the result.
English translation
Optional EN mode translates the final text into English.
Free during early launch
Get full access to every feature while we collect the first feedback.
Early launch
Download for Mac — freeUse VoiceToPrompt for free
Paid plans will appear later. Early users will get a special offer.
Future plans
Planned pricing after launch. Payments are temporarily disabled during the free early launch period.
Monthly
€6/month
VAT included where applicable
Try it without commitment. Cancel anytime.
- All features
- 3 Macs
- Cancel anytime
Save 32%
Annual
€49/year
VAT included where applicable
Best option if you use it daily.
- All features
- 3 Macs
- Best price
Lifetime
€69
VAT included where applicable
Early access price. Pay once, use it forever.
- All features
- 3 Macs
- 1 year of updates included
You pay OpenAI directly.
System requirements
Choose local recognition on Apple Silicon or cloud recognition through your own OpenAI API key.
Operating system
macOS 14.0 Sonoma or newer
Processor
Apple Silicon: M1 or newerIntel Mac is supported, but local recognition is not available.
Memory
Minimum: 8 GB RAMRecommended: 16 GB RAM or more for local Whisper models.
Disk space
From 300 MB for the app and support files. Local Whisper models download separately, starting from about 70 MB.
Recommended setup
- macOS 14 or newer
- Apple Silicon M2 / M3 / M4
- 16 GB RAM or more
- 3-5 GB free disk space for local models
Common questions
Yes. VoiceToPrompt uses local models, so after installation it can run fully in local mode without an internet connection.
During the early launch period, VoiceToPrompt is available for free. After this period ends, early users may receive an additional discount or other special terms.
In local mode, never. Everything is processed on your device through WhisperKit. In cloud mode, audio is sent to OpenAI with your personal API key. VoiceToPrompt never processes your audio on its own servers.
Interface: English, Russian, Ukrainian. Speech recognition: all 99 Whisper languages. Optimized for Russian-English and Ukrainian-English mixed speech.
Yes — 3 Macs per license. You can deactivate a Mac in Settings to free up a slot.
You only need an OpenAI API key if you want to use cloud recognition: for example, to get faster results or if your Mac does not support local models. The key is added in the app settings and used directly for your requests to OpenAI.