Turn your text data into a full length podcast or audiobook with AI!
Supports:
- ElevenLabs
- OpenAI
- Unreal Speech
Play /example/eleven_labs_intro-to-algos-asmr-lmao.mp3
This is an extract from "Introduction to Algorithms Fourth Edition" purchase the book from MIT here: https://mitpress.mit.edu/9780262046305/introduction-to-algorithms/
As it was just copied straight from the PDF there are a few syntax errors within the data that have resulted in strange generations.
when doing real world testing making sure to clean your data fire and do a few small test runs while modifying the voice settings within your given AI providers settings located in ./src/ai_provider_settings
ElevenLabs API reference https://elevenlabs.io/docs/api-reference/text-to-speech
OpenAI API reference https://platform.openai.com/docs/api-reference/audio/createSpeech
Unreal Speech API reference https://docs.unrealspeech.com/#b555f89b-526e-4212-97d3-6c9687bb2219
-
pip install -r ./requirements
-
Create an ElevenLabs/OpenAi/Unreal Speech account and retrieve an API key
-
Create a .env file and add your API key (using env.example as a template)
-
Copy and paste the text you want to turn into a podcast and put it in ./data/test_data.txt
-
Run
python ./src/determine-cost.py elevenlabs|openai|unrealspeechto work out how much it would cost to create dependent on the AI provider you're using -
Run
python ./src/determine-generation-time.py elevenlabs|openai|unrealspeechto find out an estimate time it will take to generate -
Run
python ./src/generate-podcast.py elevenlabs|openai|unrealspeechNote: Running the generation script will chunk up the files and process each of them in order to mitigate any overly long API requests that become more volatile over their run time
-
Run
python ./src/merge-podcast.pyto merge all audio files within ./output/chunks into a single file named output.mp3 -
Enjoy!
For more open source hacky projects: visit my github :)