🎙️ ListenForge

📣 What this app will do for you

This app will take a meeting note spoken live into the browser mic and turn it into written meeting notes from the live dictation (uploaded audio is not transcribed). It's your Dictation + meeting notes — built for exactly that job.

Dictation and meeting notes, privacy-first. Speak, record, or play back audio — then shape it into clean notes with speaker labels and action items.

$49 one-time.

Buy ListenForge — $49 one-time

Buy above to get instant access — the tool opens right from this page.

How it works

  • Live dictation — speak and watch words appear, using your browser's built-in speech recognition where your browser provides it.
  • Record in the browser — capture audio with your microphone and play it straight back. The recording never leaves your machine.
  • Play back an audio file — upload a recording and listen while you type or dictate. Uploaded files are not auto-transcribed (see the honest limit below).
  • Speaker labels — tag segments with a speaker name so a meeting transcript reads like a conversation, not a wall of text.
  • Action items, by rules you can read — lines containing a commitment word (will, need to, action, todo, follow up, deadline…) or a date are flagged as candidate action items. The rules are listed in the tool, and the tool never pretends it understood more than that.
  • Optional AI polish — one button tidies your notes. When AI enhancement is available it is clearly labelled as AI; otherwise you get the same deterministic tidy-up, labelled as such.

🔒 Privacy, stated plainly

Transcription runs in your browser / on your machine where supported, and ListenForge itself never sends your audio or your notes anywhere — there is no upload, no account sync, and no analytics inside the tool. Closing the page loses unsaved work, so use Download notes to keep a copy.

Two honest exceptions, both labelled when they happen:

  • Your browser's speech recognition may send audio to your browser vendor (for example, Chrome's recognizer sends audio to Google for recognition). That is your browser's behaviour, not ListenForge's — ListenForge adds no upload of its own. If that matters to you, use the recorder plus manual typing instead of live dictation.
  • The optional AI polish sends your notes text to the AI provider when you press that button and AI enhancement is available. It is never automatic, and the result is labelled as AI-enhanced. Your audio is never sent for AI polish — only the text you already typed or dictated.

What it cannot do (yet)

  • Uploaded audio files are NOT auto-transcribed. No on-device speech model is bundled with ListenForge, so an uploaded file cannot be turned into text in your browser. You listen to it and type, or dictate live while it plays.
  • Live dictation depends on your browser. Browsers without the SpeechRecognition API (including some mobile browsers) show a clear "not supported here" message instead of a fake dictation button.
  • Speaker labels are assigned by you. ListenForge does not detect who is speaking — you label each segment.
  • Action items are rule matches, not understanding. A flagged line is a candidate to confirm, and real commitments phrased without a listed cue word will be missed. Read the flagged list before trusting it.