bing_dai
Joined 12 years ago
Karma
555
Submissions
127
Member since
Aug 14
Recent submissions
Comment · 2 years ago
Could you provide an example?
Comment · 2 years ago
Project 1
— Source code: https://github.com/bingdai/summaryfeeds. The code is for Summary Feeds (https://www.summaryfeeds.com). It shows summaries of AI-related YouTube Channels.
****
Project 2 - I also built a YouTube summarizer for individual video called Summary Cat (https://www.summarycat.com). It is not open source for now. The stack is very similar to project 1.
****
And yes I like summarizing YouTube videos:)
Comment · 3 years ago
Mainly because knowledge about Christianity is the most physically accessible to me, as I am a next-door neighbour to a Theology professor of New Testament.
Comment · 3 years ago
Learning about Christianity, after being an atheist my whole life thus far.
Comment · 3 years ago
https://www.summarycat.com/
I made a YouTube summarizer, Summary Cat.
————————-
Technical Details
Goal: Generate brief and coherent summaries from YouTube video transcripts
Motivation of building this project: to save myself time watching videos that are too long.
Stack • Frontend: HTML/CSS, plus a tiny bit of Javascript for handling the button • Backend: Python/Flask • Flask served as the backend web framework, handling routes and integrating various services. • Front+Backend has about 500 lines in total • Hosting: AWS Free-tier Elastic Beanstalk • Database: AWS Free-tier RDS
APIs: • For grabbing YouTube's transcripts: I used youtube-transcript-api (https://pypi.org/project/youtube-transcript-api/) • For summarizing the transcripts: I used OpenAI's GPT-3.5-turbo-16k: https://platform.openai.com/docs/guides/gpt. • I used GPT-3.5 because GPT-4 is quite a lot more expensive (roughly 10X).
My Prompt (Super Simple!) • "please summarize the following text into a few paragraphs:" + the full transcript.
Thoughts about GPT-4 vs GPT-3.5-Turbo-16k or Summary Cat • GPT-4 was 20% better for "summary quality" • GPT-4 feels 50% faster • However, GPT-4 is about 10X as expensive as GPT-3.5
Comment · 3 years ago
Ah ha! Great point!
Comment · 3 years ago
Hi, I used this Python library (https://pypi.org/project/youtube-transcript-api/) to get transcripts. It works great.
Comment · 3 years ago
I agree that this is happening " an arms race between youtube forcing creators to make videos 8 mins long min to be able to get mid roll ads and people coming up with ways to summarize the transcript." Along the same line: I have been thinking about how my Summary Cat might mean for the content creators. How would it impact their income?
Your "use the summarizer to re-cut the videos" is fantastic!
Comment · 3 years ago
Sorry, could you please explain what you mean? I am not really quite getting it. What might happen if a Prompt Engineer get a hold of it?
Comment · 3 years ago
I think your suspicion might be correct: long videos exceeds GPT token limitation (16,385 tokens in my case of GPT-3.5-turbo-16k).
Thanks for your suggestion about how to address it.
Email: bingdai9 at gmail dot com