STORY
🤓 BTC Transcripts Review - Pitch
AUTHOR
JonasJonas
Joined 2023.07.12
PROJECT
BTC Transcript ReviewBTC Transcript Review
DATE
VOTES
sats
COMMENTS
NOSTR
View on Nostr ↗
TAGS
#webLN#ai

🤓 BTC Transcripts Review - Pitch

🤓 BTC Transcripts Review

There is a treasure trove of insights, knowledge, and ideas about technical Bitcoin concepts. 

But they’re currently locked inside audio and video talks. 

We’ve built a flywheel that uses AI to transcribe high-signal audio and video content and feeds it to Bitcoin Transcripts. However, given the technical nature of the content, AI only does so well. 

So, we introduced a seamless process for learners and contributors to the Bitcoin ecosystem to run human-in-the-loop edits to these AI reviews. 

Doing so deepens the reviewer's understanding of the content. And they also get paid for their contributions in ⚡ sats ⚡!

The long-term vision is human-corrected, AI-generated transcriptions of Bitcoin talks. 

Why? These talks become scalable, indexable, and usable for learners, builders, and the future users of Bitcoin.

🎢 Experience for Yourself

Check out our live product below!

⚒️ How We Built This

  • We built using light scrum methodology cycles of “plan, build, review, adjust”

  • All our code is for the public to play and fork on github

  • We ongoingly solicited feedback from users to ensure we’re meeting and exceeding their needs (we’re building for them after all!)

  • We used surveys and moderated usability tests to get feedback about our products

💽 Technical Stuff

  • The front end is a nextJS app

  • Whisper transcripts are generated from TSTBTC (built before July) and sent to the transcription-review-backend (started in spring).

  • Hosted postgresql

  • We use voltage to host our LND node and have Alby integration for payments.

⏩ And Then What?

Now

  • Transcripts Review: get paid to review bitcoin transcripts

Next

  • Transcripts Review: integrating a read-along editor like Scription

Later

  • Bounties programs

  • Human-supported feedback to teach future AI-generated model

🫂 Team