Skip to content

USER_GUIDE.md

Latest commit

 

History

History
121 lines (85 loc) · 3.31 KB

USER_GUIDE.md

File metadata and controls

121 lines (85 loc) · 3.31 KB
Raw

Echo User Guide

Echo is a macOS menu-bar voice dictation app that records speech, transcribes it, refines with AI, and types the result into your active application.

Installation

From Source (Developers)

brew install sox
cd echo
npm install
npm run setup    # Build local Whisper (optional)
npm start

Permissions

On first launch, Echo will guide you through:

  1. Microphone — required for recording
  2. Accessibility — required to insert text into other apps
  3. Speech Recognition — required for live transcription (macOS STT)

Recording

Method Action
fn key (hold) Hold fn to record, release to stop
fn key (double-tap) Toggle recording on/off
fn key (single-tap) Stop toggle recording
Hotkey (default ⌘⇧V) Toggle or hold-to-talk (per Settings)
Overlay click Toggle recording
Tray menu Start/stop recording
Esc Cancel recording

Settings Overview

General

  • Recording Mode — Toggle (press to start/stop) or Hold (hold keys to record)
  • Start Delay — Milliseconds before recording begins (prevents accidental triggers)
  • Silence Detection — Auto-stop after period of silence
  • STT Engine — Choose transcription backend
  • Transcription Language — Language for STT (or auto-detect)

LLM

  • Provider — Claude CLI, Codex CLI, Claude API, OpenAI API, Ollama, Llama.cpp, or None
  • Grammar Check — Second pass for grammar/punctuation
  • Voice Commands — "new line", "scratch that", etc.
  • Templates — Voice-triggered text snippets

Memory

Add vocabulary entries so Echo learns your terms:

  • Term — Correct spelling (e.g., "React")
  • Misrecognitions — What STT gets wrong (e.g., "react, re act")

Echo also auto-learns corrections after 3 consistent matches.

History

Search and re-insert past dictations from the History tab.

Voice Commands

Command Result
"new line" Line break
"new paragraph" Paragraph break
"period" .
"comma" ,
"scratch that" Cancel (skip refinement)
"undo that" Cancel (skip refinement)

Templates

Create templates triggered by voice:

  • Name: Email Signature
  • Trigger: "type my email signature"
  • Content: Best regards,\nYour Name

Troubleshooting

Problem Solution
SoX not found brew install sox
Whisper not ready Settings → build binary + download model
Accessibility denied System Settings → Privacy → Accessibility → Echo
Groq/API errors Validate API key in Settings
Ollama not running ollama serve
No speech detected Check microphone, speak closer

Copy Logs

Settings → General → Copy Logs for bug reports.

Provider Health

Settings → LLM → Provider Health shows status of all configured backends.

Keyboard Shortcuts

Key Action
⌘⇧V Record (default hotkey)
⌘⇧B Toggle overlay
Esc Cancel recording
fn (hold) Hold-to-talk

Fully Offline Setup

  1. STT Engine → Local Whisper.cpp
  2. Build Whisper binary + download model in Settings
  3. LLM Provider → Llama.cpp or None
  4. Disable window context and screenshots

See PRIVACY.md for data handling details.