Learn Build Teach Hackathon Submissions

[BekahHW]

As a reply to this post, submit your project using the following template:

## Title
### Description
{{Text for short description}}

**Link to code:**
**Link to blog post:**

The completed submission should look like this:

Building a Voice-Powered Song Search

Description

For this project, I access the microphone to power a Christmas song search using the Genius API.

Link to code: https://github.com/deepgram-devs/song-search
Link to blog post: https://blog.deepgram.com/song-search-js/

[dotAadarsh]

Building YouTXT

Transcribing YouTube Videos

YouTXT is an app that transcribes any YouTube video and provides some additional features like summary, keywords, Transcript translation, search the video, and markdown editor to edit the transcription.

Link to code: https://github.com/dotaadarsh/YouTXT
Project Live URL:: https://youtxt.streamlitapp.com/
Link to blog post: https://aadarshkannan.hashnode.dev/building-youtxt

[Manikant92]

VProfressor.ai — A 24x7 Virtual Professor for Students

An application that it helps the students and increase their speed, efficiency of learning and also get their doubts clarified with additional learnings.

VProfessor.ai contains different features designed to facilitate online learning. Users can upload a wav/mp3/ text files. Then, a Deepgram STT transcript of the audio is returned, along with a summary of the data. This includes key words and main topics, Wikipedia page links, current events from NewsAPI, and recommended YouTube videos. From here, users can either read the summary report on the website or download it as a pdf for their personal studying. Additionally, students may “ask the professor” a question, and get a quick short answer for themselves which leverages wolframalpha api’s.

Link to code: https://github.com/Manikant92/VProfessor_LBTHackathon
Link to blog post: https://medium.com/@kellamanikant/vprofressor-ai-a-24x7-virtual-professor-for-students-91b01b88fee6

[Manikant92]

Can we please know when the winner results be announced?

[Manikant92]

Yes, I found the channel. But don't see any update as of now.

[Manikant92]

@BekahHW

[Manikant92]

I found the blog post from deepgram blog now. Can you please let us know the criteria of selecting the winners please? Also, as per hackathon, there are 2 grand prizes and 5 runner up prizes, only one winner and runner up prizes are announced on the page. Please let us know criteria of selection? Thanks @BekahHW

[BekahHW]

Hey @Manikant92. To be a valid submission, you need to be a member of the LBT community. As per the guidelines, the project had to be created for this hackathon. Any work submitted for any other hackathons, had to be disclosed as per the agreed to terms and conditions.

The two winners were the two valid submissions we received, and judging was based on factors including innovation, use of Deepgram, and code submission.

[Manikant92]

Thanks @BekahHW for confirmation. This is just a feedback, I came to know about this hackathon via your monthly marketing emails, however I lately realised the requirement was to be present in discord channel as well. But i couldn't find the link to join in any of your communications either email or post about this hack. So, from next time, can you please mention the relevant links in your hackathon communications so that developers who are interested to participate can join in all sources. Thanks!

[TekGadgt]

DiscordSTT

A discord bot with per user transcription functionality.

This started as a Virtual Table Top transcription concept, to allow virtual D&D campaigns to be more accessible, but we soon realized that most folks playing D&D on VTTs were using Discord for Voice. So we expanded the scope to handle Discord voice. (I don't have a blog and didn't realize this was what was meant by a github discussion, so we actually wrote our post on our repo's discussions, linked below).

Link to code: https://github.com/TekGadgt/discordstt
Link to blog post: TekGadgt/discordstt#5

[DeepPleb]

DeepCensor

Censor swear words automatically

Using profane language is often unwelcomed. You don't hear an explicit song on the radio. A Youtube video with swears could be demonetized. Live news with profanity might even land the channel in legal trouble. But censoring audio can be a time consuming task.

Thanks to AI we can now do this automatically. DeepCensor is a piece of python code which takes in an audio file (or a live stream link) and spits out an output file with all the swears muted. This process is entirely automatic thanks to the Deepgram python SDK.

Link to code: https://github.com/DeepPleb/DeepCensor
Link to blog post: https://medium.com/@scarphaze000/deepcensor-1c402df7f270