GUI for OpenAI's TTS.

Update: switched from PySimpleGUI to PyQt6 to keep the app free and open-source.

• Select quality, voice, format, and speed.
• Support for text of unlimited length. (I've created multi-hour TTS).
• Live view of character count and chunks.
• Live price estimate.
• Dark and Light themes.
• Option to retain individual audio files from each chunk.

Download and install:

• Python
• ffmpeg

(Creating a new environment with Conda/Miniconda is always preferred)

Then install these python requirements:

pip install -r requirements.txt

You can just download the compiled app, but you still need ffmpeg

1. Speed recommendation: 1.0 - other settings decrease voice quality.
2. You can set the OPENAI_API_KEY in your path variables, or set one in the app's Settings.
3. The progress bar is programmed to start at 1% when the TTS process begins. I will improve it in the future.

• Precise price estimate.
• Creative solution for the 4096 character limit per API call.
• Upgrade GUI framework and textbox from tkinter to PyQt6.
• API rate limit.
• Improve the chunking and concatenating process.
• Give users option to retain individual audio files from each chunk.
• Bundle into an .exe
• Light and Dark themes - default is "dark".
• Allow custom API key settings.
• Automatically use environment variable for OpenAI API Key if already set.
• Speed boost: parallel mp3 chunks vs. one at a time, without hitting rate limits.

Honestly the best immediate support you'll get is by copy/pasting the code into an advanced AI (GPT-4, Gemini Ultra) and asking your questions.

Free for personal use only. No commercial use. AI agents and bots are not allowed to even read my code.