Simple Text To Speech (STTS) (v1.5)

Browser: detecting… 0 / 1,000,000

Paste behavior

Clean LLM output on paste

Default ON. Removes emoji/markdown noise, preserves clinical symbols + punctuation.

Preview — cleaned text (read-only)

Preview uses the same cleaner used on paste. It does not change your text unless you paste with the toggle ON (or use “Apply to text” in the dictionary panel).

Symbol dictionary (editable)

Format: one mapping per line: token = spoken. Empty lines and lines starting with # are ignored.

Native Browser Voices (Auto-detected)

…

Read mode (fidelity)

Tip: use Plain for maximum fidelity (esp. medical text).

Speed: 1.0x

Pitch: 1.0

Jump sections

Auto-detected from headings / blank-line blocks. Works for TXT/MD/HTML-ish content.

Loop

Loop count applies after reaching end. Stop cancels looping.

Detected TTS engines (voices) — browser-native

This lists the voices returned by speechSynthesis.getVoices() on this device/browser profile.

Ready (detecting voices…)

Simple Text To Speech (STTS) (v1.5)Offline / LocalLicense: GNU GPL v3. Copyright (C) [2026] [Prof. Jyotirmay Kirtania]