Hands-Free AI.
Whisper Speech-to-Text, Voice Workflows.
Dictate to your AI assistant using Whisper transcription. Hands-free task execution, voice-first content creation, and multilingual voice input - with audio processed locally when privacy matters.
The reality
Most people think faster than they type. Voice input eliminates the bottleneck between thought and text - but most voice AI tools are either basic dictation with poor accuracy or cloud services that transmit your audio to remote servers. The combination of good accuracy and local processing has been missing.
Whisper changes this. Integrated with Skales, you get accurate multilingual transcription feeding directly into an AI agent that can act on what you said. Not just transcription - execution. Dictate a task and it gets done, not just written down.
How Skales handles voice input
Whisper transcription into AI execution - hands-free from start to finish.
Whisper speech-to-text
Skales integrates with OpenAI Whisper for accurate, fast transcription. Speak your request naturally and Skales converts it to text, processes it, and responds. Works in multiple languages and handles accents and speaking styles reliably.
Hands-free task execution
Dictate emails, summarise documents, set reminders, and run workflows without touching your keyboard. Skales listens, understands the task, and acts - useful when your hands are occupied with something else.
Voice to structured output
Dictate rough thoughts and Skales structures them. Speak meeting notes in stream of consciousness and get a formatted action list. Dictate a draft and receive a polished document. Voice input does not mean unstructured output.
Multilingual voice support
Whisper handles over 50 languages. Dictate in French, German, Spanish, Japanese, Portuguese, or your native language and receive accurate transcription. No need to speak in English to use voice input.
Voice-first content creation
Writers, podcasters, content creators, and professionals who think better out loud can use voice as their primary input method. Dictate a first draft and use Skales to shape, edit, and refine it without switching to a separate transcription service.
Audio stays on your machine
When using local Whisper processing, your audio never leaves your device. No cloud transcription service receives your voice recordings. For sensitive meetings, confidential dictation, or personal privacy, this matters.
Example: Voice-first morning
You say: "[Speaking] Ok so I had a call with the client and they want the proposal by Friday, they also mentioned the budget has moved to about eighty thousand, and they need a separate section on implementation timeline."
Skales: Structured meeting note: proposal deadline Friday, budget updated to $80k, scope addition flagged. Action items extracted and ready to copy into your project tracker.
You say: "[Speaking] Write an email to Tom telling him the Tuesday meeting needs to move to Thursday, same time, because I have a conflict. Keep it short."
Skales: Draft email ready in seconds. Clear, professional, asks Tom to confirm the new time.
You say: "[Speaking] Remind me in two hours to check on the build status and also add review the Q3 report to my task list for this afternoon."
Skales: Two-hour reminder set. Task added to your afternoon list. No keyboard required.
“I dictate while I walk between meetings. By the time I sit down, the emails are drafted and the tasks are logged.”
Free for personal use. Windows and macOS. No account required.
Also see: Accessibility · Multilingual AI · Download