Talk to your podcast transcripts.
To install and run this project, follow these steps:
- Install Mamba in order to install reqired dependencies.
- Clone the repository:
git clone https://github.com/mgogov/openpod.git
- Navigate to the project directory:
cd openpod
- Create a mamba environment from
requirements.txt
and activate it:
mamba create -n openpod
mamba activate openpod
pip install -r requirements.txt
Currently only OpenAI's gpt-3.5-turbo
and gpt-4
are available.
Provide a .env
with the following content in the home dir:
OPENAI_API_KEY=<your_API_key>
$ python openpod.py usage: openpod [-h] [--chat] [--reindex] [--llm {openai-gpt-3.5-turbo,openai-gpt-4}] [--eval] [--eval-id EVAL_ID]
[--eval-reset-db]
OpenPod command line interface.
options:
-h, --help show this help message and exit
--chat Chat with the podcast
--reindex Force reindexing of the podcast data
--llm {openai-gpt-3.5-turbo,openai-gpt-4}
Specify the LLM to use
--eval Evaluate how well the LLM is answering the benchmark questions
--eval-id EVAL_ID The ID to be used for the evaluation
--eval-reset-db Whether to reset the evaluation DB
Contributions are welcome! If you would like to contribute to this project, just open a pull request.
This project is licensed under the Apache2 License.
If you have any questions or suggestions, send me an email (check out my profile).