Settings Guide

Configure MurmurTone to work exactly how you want.

Access settings by right-clicking the MurmurTone icon in your system tray and selecting "Settings".

General Settings

Speech Model

Choose the model that best fits your needs. See the full Model Guide for detailed comparisons.

Model	Speed	Accuracy	RAM Usage	Best For
`Quick`	Fastest	Good	~1 GB	Quick notes, chat
`Standard`	Fast	Better	~1.5 GB	Casual dictation
`Recommended`	Moderate	Great	~2.5 GB	General work
`Professional`	Slower	Excellent	~5 GB	Important documents
`Studio`	Slowest	Best	~10 GB	Professional, accents

Recommendation Start with Recommended for the best balance of speed and accuracy. Use Professional or Studio if you need higher accuracy and have the RAM/VRAM.

Processing Mode

Select how MurmurTone uses your hardware:

Mode	Description
Auto	Automatically selects best option based on your hardware
CPU	Uses only CPU (works on any machine)
GPU-Balanced	Uses GPU with balanced settings (faster, requires NVIDIA GPU)
GPU-Quality	Uses GPU with quality settings (fastest, requires NVIDIA GPU)

Language

Choose your transcription language:

English (en) - Optimized for English only (faster, more accurate)
Auto-detect - Automatically detects language from 18+ supported languages
Specific language - Choose a specific language (Spanish, French, German, etc.)

Hotkey

Set your push-to-talk hotkey:

Click the "Set Hotkey" button
Press your desired key combination (e.g., Ctrl+Shift+Space)
The new hotkey is saved automatically

Avoid conflicts Choose a hotkey that isn't used by other applications. Avoid common shortcuts like Ctrl+C or Alt+Tab.

Audio Settings

Input Device

Select which microphone to use for recording. Options include:

System Default - Uses your Windows default recording device
Specific device - Choose a specific microphone from the dropdown

Noise Gate

Filter out background noise:

Enable/Disable - Toggle noise gate on or off
Threshold - Set the dB level below which audio is ignored (-60 to -20 dB)

Lower values (e.g., -50 dB) are more sensitive and let more sound through. Higher values (e.g., -30 dB) filter out more background noise but may cut off quiet speech.

Audio Feedback

Toggle individual sound effects:

Recording start - Beep when recording begins
Processing - Sound while transcribing
Success - Chime when transcription completes
Error - Alert when something goes wrong
Command - Sound when voice command is recognized

Advanced Features

Custom Vocabulary

Add technical terms, names, and jargon to improve recognition:

Go to the "Custom Vocabulary" tab
Enter words or phrases, one per line
Click "Save"

Examples: Company names, product names, technical acronyms, people's names.

Custom Dictionary

Create text replacement shortcuts:

Go to the "Custom Dictionary" tab
Add trigger phrases and their replacements
Click "Save"

Example: "my email" → "yourname@company.com"

Translation Mode

Speak in one language, output in English:

Enable "Translation Mode" in settings
Select your source language (or auto-detect)
Speak in your selected language
Transcription appears in English

AI Text Cleanup

Automatically clean up transcriptions using local AI. Enable in Settings → Advanced → AI Cleanup.

Preview Window

Show a floating preview of your transcription:

Enable/Disable - Toggle preview window
Theme - Choose light or dark mode
Position - Set where the preview appears on screen

Settings File Location

Settings are stored at:

%APPDATA%\MurmurTone\config.json

You can back up or transfer this file to preserve your settings across installations.