A kinder keyboard for hard messages

Say what you mean, with the right tone.

EmapthyAi helps you turn a rough draft into clear, human communication. Choose a style, rewrite the text, or hear the result in your own voice.

What it does

EmapthyAi lives where you already write. It keeps your original message intact while giving you a deliberate choice about how the message should land.

Rewrite the full draft

Read the current text box on every action, then ask the configured language model for a complete rewrite in the selected style.

Pick a persona

Switch between Corporate, Personable, Warm, and Empathy to match the relationship and moment.

Speak it naturally

Send the rewritten text to ElevenLabs and play it back with your saved voice fingerprint and persona-specific delivery.

How it works

1

Write or dictate

Enter text in the active composer, or use Voice input to transcribe what you want to say.

2

Choose a style

Open the persona menu and select the tone that fits: Corporate, Personable, Warm, or Empathy.

3

Rewrite or Speak

Rewrite replaces the draft with the complete model response. Speak rewrites it first, then sends that complete text to ElevenLabs.

4

Review before sending

Keep the original, use the rewrite, play either version, and decide what to send. EmapthyAi does not send a message for you.

Desktop recording · September 12, 2026

See the keyboard in action

This short recording shows voice setup, the test area, persona selection, rewrite, and voice playback on iPhone.

The original recording is included with this deployment so the demo remains available alongside the explanation.

Run it locally

The API is designed for local development first. Start it with Node 22+, Postgres, and server-side provider keys; clients never contain model-provider secrets.

cd website
cp .env.example .env
npm test
npm run start:api

See the full README for server configuration, iOS and Android setup, browser-extension behavior, and verification commands.