Pummy.TTS — Text-to-Speech Made by a Medical Student, for Medical Students

I made Pummy.TTS because I saw a need for free study tools that simply work and fit the way medical students use Anki. As a medical student myself, I wanted an easier way to listen to my cards, follow along with the text, and hear medical terminology without having every source citation read aloud.

Pummy.TTS is my attempt to make that experience easier. It reads your cards aloud, highlights words as they are spoken, and gives you control over what gets read and when.

It is free to use, with no subscription, account, or paid speech API required.

WHAT YOU CAN DO

• Listen to the front and back of your cards, with optional automatic playback and separate controls for each side. • Choose whether flipping a card reads just the answer or repeats the front first. • Follow spoken words with highlighting in a color you choose. • Skip common citations, Source lines, slide references, and textbook page references while keeping them visible on your cards. • Use a searchable, editable dictionary with 626 medical abbreviation and pronunciation entries covering anatomy, physiology, genetics, biochemistry, and clinical terminology. • Adjust cloze reading: hear the revealed answer, the full card, or nothing. Choose how hidden text is handled on the question side. • Select supported installed system voices, adjust reading speed, and preview a voice before studying. • Optionally use the included English Kokoro voices: Heart, Bella, Michael, and Fenrir. Speech is generated locally on your computer. • With Kokoro selected, optionally prepare audio for existing decks or newly added decks, with a progress bar and cancel controls. New-deck preparation is off by default. • Use a resizable settings window that follows your system's light or dark appearance, with additional theme choices.

** Kokoro takes a while to download and doesn't handle speeds higher then 1x well. I will update with a better version if I find one**

GETTING STARTED

Install the add-on, restart Anki, and open Tools → Pummy.TTS → Settings. Choose your speech engine and voice, try the voice preview, and set your reading preferences. Reading speed starts at 1×. System voices are the default. Kokoro is optional and includes its own local model and voices. Its first playback may take longer while the model loads. Above 1×, it prepares the full recording before playback to reduce pauses between phrases. Preparing deck audio in advance can reduce the wait for matching voice and speed settings.

• Replay: Alt+R on Windows / Option+R on macOS • Stop: Alt+S on Windows / Option+S on macOS

COMPATIBILITY AND CURRENT LIMITS

Configured to allow Anki Desktop 26.05, and tested against 26.09.3 on Windows. macOS support is included, but has not yet been tested on a physical Mac. Desktop add-ons do not run in AnkiWeb or mobile apps.

Kokoro currently supports English. Available system voices depend on what your operating system exposes to third-party apps. Pronunciation and citation filtering can vary, so the text preview and editable dictionary let you tailor the reader to your course. Image-occlusion cards are skipped by default.

Prepared audio is stored in a local cache of up to 512 MB per profile. It is not added to your cards or synced to other devices. Your notes, card templates, and scheduling stay unchanged.

I built Pummy.TTS to make my own studying easier, and I am sharing it for free in the hope that it helps yours too. If you notice any bugs please let me know email: Daniel.Dehkordi1@gmail.com.

Version 1.4.1

Includes optional Kokoro voices, deck audio preparation, customizable highlighting, medical pronunciations, autoplay and cloze controls, and improvements to increased-speed playback and settings resizing.

Open-source speech components and their license notices are included with the add-on.