Deployed on https://replit.com/@bholagabbar/ai-insights-generator . Fork the repl and it should be good to go. The configuration can be adjusted for a business specific usecase or to have the LLM act as a subject matter expert.
This script accepts large input data, such as customer research calls or transcripts from various sources (eg from fireflies ai), and generates insights and recommendations based on the provided data. It can handle input data from the following sources: HTTP(S) URLs, local files, or user input prompts. The script utilizes the OpenAI API to interact with the gpt-3.5-turbo-16k model and generate these insights and recommendations.
- Customer Research Calls: Derive insights and focus areas from customer research calls or transcripts to drive startup success.
- Market Research: Analyze surveys, reports, and competitor data to generate insights for informed business decisions.
- User Feedback Analysis: Process support tickets, online reviews, and social media feedback to improve products/services.
- Strategy Development: Utilize industry reports, trends, and expert opinions to inform strategic decision-making.
-
Clone the repository:
git clone https://github.com/bholagabbar/ai-insights-generator.git
-
Install the required dependencies:
pip install -r requirements.txt
-
Run the script optionally research document file as an argument, or a link to a a transcript. If you don't have either, you can run it as is, and it will prompt you to paste the text or URL:
python main.py
or
python main.py https://jsonkeeper.com/b/WKIV
or
python main.py your_research_document.txt
-
The insights and recommendations are printed on the screen, as well as saved to an output file
-
git config --local user.name ""git config --local user.email ""- Store token in
.githubinfoalong with potentially other Git information git push https://github.com/your-username/fireflies-transcript-splitter.gitand enter the password/token when prompted
This project is licensed under the MIT License.