HNHacker News
TopNewBestAskShowJobs

bing_dai

555 karma · joined August 14, 2014

Web: www.summarycat.com

Email: bingdai9 at gmail dot com

submissionscomments
bing_dai··on In many cases we need something less like AI and more like a basic algorithm (2023)
Could you provide an example?
bing_dai··on Ask HN: What have you built with LLMs?
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:)

bing_dai··on Ask HN: What are you passionate about at the moment?
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.
bing_dai··on Ask HN: What are you passionate about at the moment?
Learning about Christianity, after being an atheist my whole life thus far.
bing_dai··on Ask HN: Show me your half baked project
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

bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
Ah ha! Great point!
bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
Hi, I used this Python library (https://pypi.org/project/youtube-transcript-api/) to get transcripts. It works great.
bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
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!

bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
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?
bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
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.

bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
If you don't mind, could you share an example of such long video? I'd love to debug it.

As for non-Youtube URL: I indeed do not handle that error right now :D. Thanks so much for bringing it up!

bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
You are right. The site doesn't handle non-English videos, but that is on the roadmap. Thanks so much!
bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
More than happy to share!

----------- My prompt is super simple. It is "please summarize the following text into a few paragraphs:" + the full transcript. -----------

Seriously that's it!

bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
Hello, GPT-4 is not worth it in my experience so far!

I would say, GPT-4 is - 20% better at "summary quality" - feels 50% faster - BUT, 10X as expensive.

So using GPT-3.5 was the right choice for me at this point.

bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
Thanks for the question... Any sample videos you are thinking of?

Summary Cat doesn't work for videos where there's not a lot of speaking. I am hoping to build a bit more on text-rich videos first, so I do not plan to do actual video (image) processing any time soon.

I will keep that in mind!!

bing_dai··on Show HN: Summary Cat, a YouTube Video Summary Generator
Hello! Thanks for the question. I do not myself restrict tokens/session. The model I am using GPT-3.5-Turbo-16k (https://platform.openai.com/docs/models/gpt-3-5), allows max 16,385 tokens in total per input/output.

So far, I found that each ~10 minute video uses around 1000 tokens. It costs me about 3 cents to summarize, which is not too bad as I don't have many users, and users haven't been requesting summaries for super long videos yet.

If this sites gets a lot of interest, I might start restricting something :)..

GPT 3.5 Pricing: https://openai.com/pricing

bing_dai··on Ask HN: What is your spiritual practice?
"Our constructed definition of 'good' comes from religion."

I am wondering what makes you think that?

bing_dai··on No meetings, no deadlines, no full-time employees
There are people who are perfectly happy with about $250K of take-home pay a year if that means they have plenty of freedom to do other things. In fact, I bet many people in the US will be happy with $30K a year, not $250K.
bing_dai··on To the brain, reading computer code is not the same as reading language
Fascinating! Any chance I could know which school it is?

I have worked in both computer science and classical music, and found many crossover talents. Two previous Director of Music at my school (UBC in Canada) in the last few decades are both musicians and engineers/physicists.

bing_dai··on The global fertility rate is falling
On the other hand: there must be people who decided they wanted children in high school, but later regret it.

I find it unfair that the society often consider people being "too young; may change mind later" when they say they do not want children; but we don't apply the same standards to young people who want children.

bing_dai··on The global fertility rate is falling
May I suggest you consider fostering a child? Or spending more time as an educator of some sort (through teaching, mentoring, etc.)? There are definitely ways for you to give love to a younger generation that does not involve your daughters decision to have children or not.
bing_dai··on The global fertility rate is falling
"> My ancestors thought it was okay to have kids during times in which many scholars where certain the apocalypse was due in the next 25 years, during periods in which the plague killed more than half of the population, during the 30 years war and in a time when there was a realistic chance of nuclear annihilation in the next 10 years. I am glad they did."

In my opinion, your ancestors had children largely because they had intercourse without accessing to birth control, not because they made a conscious decision to have children despite the circumstances.

bing_dai··on The global fertility rate is falling
"Society can only support you in old age if there are enough younger people around."

I do not agree with that. In my opinion, technology advancements (say, driverless cars) will make reduce the amount of young people needed to support for older people.

bing_dai··on The global fertility rate is falling
One of the most fascinating books I read is Pricing the Priceless Children by Viviana A. Zelizer (https://www.amazon.ca/dp/0691034591).

The central theme is that modern children are "economically useless" and "emotionally priceless".

bing_dai··on Marc Andreessen interview
A related tip: make sure to be kind to the assistants.

- Build rapport with them. - Shake their hands. - Talk to them with respect.

Marc Andreessen probably has literally hundreds of tasks on his desk a day.

Whether your task (say, a fundraising document you need him to sign today) would become his priority can 90% depend on whether you and his assistant have a good relationship or not.

bing_dai··on Paul Graham’s thoughts, discoveries, and tricks on parenthood
Please help me with the logic here:

"It concerns something I didn't figure out until I was older, that children are brought into existence w/o their consent.

Now I don't believe this is* (edit: fixed) an evil or unethical act; it's simply an unavoidable facet of reality. However, it creates a debt to children so massive, that it can't be fully repaid in a lifetime."

I wonder, why does "my kid didn't consent to be born" mean "I owe him/her?"

Is the implication that being alive is a bad thing?

bing_dai··on Launch HN: Art in Res (YC W20) – Buy art directly from artists
Congratulations, John! What a great initiative. I am in the artistic community (classical music) as well, so it's very inspiring to see you launching a project for the community.

I have a suggestion and a question.

Suggestion: your site does not have an About page of some sort to talk about the story behind the creation of the company and the experience of the co-founders. You have truly remarkable story, as I read in this post. So please include it to your site!

Question: I am curious about how COVID-19 has impacted the art community? My classical musician friends are certainly suffering financially and mentally because all of the concerts are suspended. I wonder how are artists doing in the pandemic.

Thank you and keep up with the great work!! Wishing you all the success.

bing_dai··on Show HN: Live Program Notes for Schubert Sonata in B-Flat
This is awe-inspiring. Thank you! It is so rare to see people who appreciate the late works of Schubert - even young professional musicians often find them too long and not "flashy" enough, which is partly why we rarely see them being played at any piano competitions.

I can see that Live Program Notes may be part of the future of classical music. Wishing you all the success!

bing_dai··on Show HN: Covid Help Center for Arts Professionals
Dear all,

Thank you for checking out the site! I started www.covidhelpcenter.com to help arts professionals and organizations find COVID-19 emergency relief funds.

I used to run a popular classical music concert series in Vancouver, Canada up until the outbreak, so I have been collecting such information to help myself and others.

I would sincerely appreciate any comments or feedbacks!

bing_dai··on High-end house rents in Vancouver plunge up to 20% as inventory spikes: experts
I just voted!
Page 1 of 3Next →