A text-to-speech application for improved reading comprehension - it's a screen reader :)
This Python application allows you to convert highlighted text to speech using OpenAI's Text-to-Speech API. It works by capturing the text held most recently in the Clipboard when a specific hotkey is pressed, converting it to speech, and then playing the audio.
I find it easier to comprehend text when I read and listen to it at the same time. I'll often listen to an audiobook and follow along in the book at the same time. I'd recommend trying it and seeing if it works well for you too!
Why do I do this? I am dyslexic and when I was diagnosed at the ripe old age of 18, it was suggested that exactly this method would help me and it did. I find the intonation in the audio version of a text helps me get the meaning first time, reading less tiring this way and I comprehend more from the text the first time I read it.
However, as a user, I feel I'm getting a subpar experience.
- Despite major AI advancements, OS system Text-to-Speech accessibility tool are still robotic sounding. This makes it less easy to process than more natural-sounding AI -generated speech.
- The right click interface kind of sucks. It doesn't work on all applications.
So, I created this terminal application to play any text that I highlighted, so I can have any text on my laptop read back to me. 🍰
The above reasons lend themselves to why NotebookLM podcasts are so great - making written information available to other parts of the brain and diversifying the formats of information makes it easier to comprehend. It goes to show that designing for accessibility sets to benefit us all.
Given that I work in AI, I depend on keeping up with research papers, which are typically dense, jargon heavy and hard-to-follow, even if you're not dyslexic! Any tools that make texts easier to follow and understand (more accessible!) make my life a lot easier.
- Capture copied text from any application
- Convert text to speech using OpenAI's TTS-1 model
- Play generated speech through your computer's audio output
- Easy-to-use hotkey activation (F8 to run TTS and play, F9 to pause/play on macOS)
Before you begin, ensure you have met the following requirements:
- Python 3.6 or higher installed on your system
- An OpenAI API key (you can get one from OpenAI's website)
-
Clone this repository or download the script.
-
Install the required Python packages:
pip install pyperclip pynput pygame openai python-dotenv -
Create a
.envfile in the same directory as the script and add your OpenAI API key:OPENAI_API_KEY=your_api_key_here
-
Run the script:
python main.py -
The application will start and wait for the hotkey to be pressed.
-
In any application, highlight and copy the text you want to convert to speech (Cmd+C on MacOS).
-
Press F8 (on macOS) to activate the text-to-speech conversion.
-
The highlighted text will be captured, converted to speech, and played through your computer's audio output.
-
Pause/resume the audio with F9
-
To exit the application, press the Esc key.
- To change the hotkey, modify the
HOTKEYvariable in the script. - To use a different voice, change the
voiceparameter in thetext_to_speechfunction. Available options are "alloy", "echo", "fable", "onyx", "nova", and "shimmer". - To adjust the speed of audio playback, modify the
speedparameter in thetext_to_speechfunction. The default value is 1.0.
- If you encounter issues with audio playback, ensure that pygame is properly initialized and that your system's audio is working correctly.
- If the text capture isn't working, make sure the script has the necessary permissions to simulate key presses and access the clipboard.
Contributions to this project are welcome. Please feel free to submit a Pull Request.
This project is open source and available under the MIT License.
- This project uses OpenAI's API for text-to-speech conversion.
- The audio playback is handled using the pygame library.