AI app for writers · no coding needed
On-device voice-to-text writing console
Reduce the number of rented dictation tools while keeping dictated text, style data and voice commands on the writer's own machine.
Made for: Writers and knowledge workers who dictate into desktop apps on their own computer
What it does for you
The problem
Dictation tools split speech-to-text, cleanup, research and task actions across several subscriptions, and most send audio to a cloud service.
What it gives you
Reviewed, source-linked dictated text and voice-triggered actions
What you give it
Microphone audiocustom termsapp contextthe writer's own style samples
Build your own version of AI dication by Snaply, Voicetypr and more
One app with what these 8 AI tools do, yours to keep and change: AI dication by Snaply, Voicetypr, VibeSonic, Stet, Yap, GHOSTYPE, OASIS 1 Ring, Clippy, but on Steroids.
Everything these tools do, in one app
- On-device speech-to-text Converts your speech into text using models that run locally on your computer.Found in AI dication by Snaply, Voicetypr, VibeSonic and 4 more
- System-wide dictation Lets you dictate into any app or website where you can type.Found in AI dication by Snaply, Voicetypr, VibeSonic and 3 more
- Hotkey activation Starts and stops dictation with a keyboard shortcut.Found in AI dication by Snaply, Yap
- Automatic grammar and punctuation Adds or adjusts grammar and punctuation while transcribing.Found in AI dication by Snaply
- Low-latency transcription Produces text with minimal delay so it appears almost as you speak.Found in AI dication by Snaply, Voicetypr, Clippy, but on Steroids
- Minimal interface Keeps the app visually simple and out of the way while you work.Found in Voicetypr, Stet
- Voice-driven notes and tasks Creates notes and tasks using voice commands.Found in VibeSonic
- Custom dictionaries Lets you add names, code, or special terms to improve recognition.Found in VibeSonic
- Web research assistant Returns cited answers from the web without leaving your workflow.Found in VibeSonic
- Bring-your-own API key Lets you plug in your own API key to use a cloud provider you choose.Found in VibeSonic, Stet
- Two-mode output Adjusts how much the text is refined depending on whether it is for a person or an AI.Found in Stet
- Clipboard preservation Keeps your copied content intact by restoring the clipboard after inserting dictated text.Found in Yap
- Small resource footprint Uses very little storage and memory so it runs lightly on your machine.Found in Yap
- App-aware sending Detects the active app and sends the text with the right command automatically.Found in GHOSTYPE
- App-specific tone profiles Switches tone and formatting based on which app you are using.Found in GHOSTYPE
- Local style learning Learns your writing style from your computer so output matches your voice.Found in GHOSTYPE
- Voice-triggered actions Performs tasks like creating tickets or editing calendar entries from voice commands.Found in Clippy, but on Steroids
- Whisper-friendly microphone Captures quiet speech so you can dictate in shared spaces.Found in OASIS 1 Ring
How it works, step by step
- Transcribe speech on-device with a local model
- Dictate into any app or website with a text field
- Start and stop dictation with a keyboard shortcut
- Add grammar and punctuation during transcription
- Produce text with low latency as the writer speaks
- Keep the interface minimal and out of the way
- Create notes and tasks from voice commands
- Maintain custom dictionaries for names, code and special terms
- Return cited web answers inside the workflow
- Accept a bring-your-own API key for a chosen cloud provider
- Switch between person-facing and AI-facing output modes
- Restore the clipboard after inserting dictated text
- Keep storage and memory use small
- Detect the active app and send text with the right command
- Apply app-specific tone and formatting profiles
- Learn the writer's style from local samples
- Perform voice-triggered actions such as creating tickets or editing calendar entries
- Capture quiet speech for use in shared spaces
- Compare the reviewed result with the recorded baseline and value assumptions
- Capture corrections and named-owner approval before consequential use
- Export a versioned reviewed dictation record with source references and unresolved questions
Build it yourself with your AI system
Build this app yourself, no coding needed
Start with a quick version you can try in a few minutes. Like it? Then build the full app by copying and pasting our step-by-step instructions: everything is prepared for you.
Sign in to see how to build it yourself
Build a quick version to try, or get the full app pack for On-device voice-to-text writing console with the step-by-step building instructions. You don't need any technical skills: you copy, paste and answer a few questions. Both are included in the membership.
4 Have it built for you days to a few weeks
Rather not do it yourself, or want it fully tailored to your data, your way of working and your brand? Nexibeo builds On-device voice-to-text writing console with you.
What's in the app pack
Included in the Complete AI Training membership.
- The building instructions your AI follows, step by step
- The questions your AI will ask you about your business before it starts
- A clickable demo you can open in your browser, to see how it should work
- A detailed blueprint of the screens, the information it keeps and the checks it runs
Become a member to get the app packAlready a member? Sign in
The files, for the technically curious
- START-HERE.mdHow to build it with your own AI (read first)3 KB
- README.mdOverview and links4 KB
- questions.mdQuestions to answer before you build3 KB
- prompt-cloudflare.mdThe full build prompt, hosted on Cloudflare26 KB
- prompt-vps.mdThe same build on your own server (Docker)26 KB
- spec.jsonData model, API, AI pipeline, acceptance criteria13 KB
- demo/index.htmlThe working demo on sample data197 KB
Questions
Do I need to know how to code?
No. You copy and paste the prompts on this page into ChatGPT or Claude, and the AI does the building. When it asks you something, you answer in your own words.
What does it cost?
The quick version, the app pack and the step-by-step instructions are for members: you pay the membership price, not a price per app (see the plans). Building the full app uses your own ChatGPT or Claude subscription. Putting it online is often cheap or no cost at the start, and your AI tells you before anything costs money.
How long does it take?
The quick version: about two minutes. The real app: an afternoon for a first version you can use, longer if you want every feature.
Can I change it to fit my business?
Yes. Tell your AI what to change in plain words, like “add a column for the price” or “use our logo and colours”. Or have Nexibeo build and customise it for you.
More detailsHow the AI works, safeguards and what to build first
Reduce the number of rented dictation tools while keeping dictated text, style data and voice commands on the writer's own machine. For writers and knowledge workers who dictate into desktop apps on their own computer, convert microphone audio, custom terms, app context and the writer's own style samples into reviewed, source-linked dictated text and voice-triggered actions. The benefit is a testable hypothesis, measured through accepted dictated words per hour and correction rate after insertion; do not assume that AI output alone produces business value.
Confirm the buyer's problem and scope, collect microphone audio, custom terms, app context and the writer's own style samples, then follow this sequence: 1. Transcribe speech on-device with a local model. 2. Add grammar and punctuation during transcription. 3. Apply app-specific tone and formatting profiles. 4. Detect the active app and send text with the right command. 5. Perform voice-triggered actions such as creating tickets or editing calendar entries. Resolve uncertain cases with the writer, approve reviewed, source-linked dictated text and voice-triggered actions, and measure accepted dictated words per hour and correction rate after insertion against a documented baseline.
How the AI works
Use AI to interpret permitted inputs, suggest structured mappings and generate candidate outputs for the three stated task modules. Use deterministic code for arithmetic, schema validation, hard constraints and reproducible tests. Review source-linked explanations and uncertainty before accepting results. Local speech models run on the writer's machine; cloud providers are optional and only through the writer's own API key. Final wording, factual accuracy and any external action remain the writer's. A model suggestion is never a verified fact, professional decision or authorization to act.
Safeguards
Preserve writer voice, source attribution, quotation accuracy and usage permissions. Writers approve substantive changes and external action scope. One desktop operating system and one local speech model; final wording and external actions remain the writer's. Keep all consequential actions under authorized human control and do not fabricate missing inputs, permissions, professional judgments or market evidence.
What to build first
Pilot scope: One desktop operating system and one local speech model; final wording and external actions remain the writer's. Implement one approved input format, a bounded representative case set and the first two task modules: transcribe speech on-device with a local model; dictate into any app or website with a text field. Support the third module with operator review: add grammar and punctuation during transcription. Include source references, corrections, basic device access, approval states, export and value measurement. Use managed operator assistance for unresolved exceptions. The cost estimate covers this narrow prototype, not unrestricted multi-tenant scale, complex production integrations, specialist certification or physical operations.
What it can connect to
Writer-owned documents, authorized meeting audio and permitted research sources. Desktop text fields, clipboard, calendar and ticket systems. Start with file exchange and validate destination specifications before promising direct publishing. Start with authorized file exchange. Validate current provider access, usage rights and schema behavior before promising a connector.
The screens in detail
Primary screens: Dictation overlay, Style and dictionary settings, Review and history console. Use a small always-available overlay for recording state, a settings panel for hotkeys, dictionaries, tone profiles and API keys, and a console listing past dictations with source audio, inserted text, corrections and linked actions. Let users compare raw and refined text side by side. Display draft, corrected and approved states. Provide a per-app profile view with the active tone and send command. Make the task-specific outcome reviewed, source-linked dictated text and voice-triggered actions visible beside its evidence, review state and value baseline.
