Summ uses ChatGPT to provide intelligent question-answering and search capabilities across user transcripts!
Easily surface insights and summarize facts across various dimensions such as department, industry, and role. With the help of natural language processing, the tool can understand and respond to complex questions and queries, making it easy for users to find the information they need.
A tool by @markiewagner and @yasyf.
You'll need an instance of Redis Stack running.
$ brew install --cask redis-stack/redis-stack/redis-stack-server
$ brew install yasyf/summ/redis-stack
$ brew services start yasyf/summ/redis-stackYou'll also need to set three environment variables: OPENAI_API_KEY, PINECONE_API_KEY, and PINECONE_ENVIRONMENT.
The easiest installation uses pip:
$ pip install summIf you prefer to use brew:
$ brew install yasyf/summ/summn.b summ requires Python 3.10+.
You can confirm that summ installed properly by running the built-in example.
$ summ-exampleFirst, create a new project with:
$ summ init /path/to/project
$ cd /path/to/project
An example implementation can now be found at /path/to/project/__init__.py.
As you can see, you don't need to do any configuration to start using summ. We simply use summ.Pipeline.default and pass a path to a directory with text files.
However, the tool works much better when users are tagged. In order to do so, you need to specify two things:
- The categories of tags (and the tags within each category).
- A prompt directing how to apply the tags of a given category.
You can see an example of this at summ/examples/otter.
Check out the summ/examples directory for some samples, or dive into the full docs at summ.readthedocs.io.
summ is distributed under the terms of the AGPL 3.0 license.