Lexicon

Say your company name. Watch it get fixed.

A personal lexicon for voice-to-agents. It maps what speech-to-text actually writes (Ashler, Cooper Nettie's, sass) back to what you meant (Ashlr.AI, Kubernetes, SaaS) before an agent sees the prompt. This page runs the real matcher in your browser.

1 What STT heard

Dictation uses your browser's own speech recognizer. In Chrome that is a cloud service: it streams your microphone audio to Google. The transcript it returns is then corrected here, on this page.

2 What the agent gets

3 Your lexicon

Add a term

Export for your dictation app


      

    

  


  

    

      

        
Install

        
Pick one. Each takes about a minute. Afterwards nothing phones home; the only download the tool ever makes is the whisper model for lexicon voice.

      


      
    
        
  1. 
          Script (macOS, Linux)
          
    
            curl -fsSL https://ashlrai.github.io/lexicon/install.sh | sh
            
          
    
        
    2. 
        
  2. 
          Homebrew
          
    
            brew install ashlrai/tap/lexicon
            
          
    
        
    3. 
        
  3. 
          npm (Node 20+)
          
    
            npm i -g @ashlr/lexicon
            
          
    
        
    4. 
      


      

        Then run lexicon setup. It seeds your lexicon with your name and company, harvests the current repo, registers the MCP server and hooks in Claude Code, Codex, Cursor and the other clients it finds, and exports to your dictation app. The script runs it for you.
      


      

        
Downloads

        
From the latest release. Checksums are in SHA256SUMS next to them.

        

          Chrome, Edge and Brave extension .zip
          Firefox extension .zip
          LexiconBar for macOS .app.zip
        

        

          Extensions load unpacked from chrome://extensions or about:debugging.
          LexiconBar is ad-hoc signed: right-click, Open on first launch.
          Walkthrough in docs/QUICKSTART.md.