Introduction
For a while now, some ideas have been running through my brain. I've basically be finding ways to ease myself of some activities in my day-to-day living by automating them. With Deepgram, I feel some of this issues will be solved. For the record; Deepgram is the first speech recognition technology am making use of.
All I want, is to see how I can use Deepgram to solve some of the issues causing a boring lifestyle for me especially in academics.
My Deepgram Use-Case
In my freshman's year in school, we were offering general courses which warrants almost all students in their first year to compulsorily take. Sometimes coming late just simply means you should not expect getting a seat unless a friend kept one for you. As if that's not enough, Lecturer's come to the lecture theatre speaking in tiny voices (also known as bedroom voice). Oh! I didn't give you the shocker, whenever I have class for 9am in the morning, I must leave my hostel as early as 5:30am and this still doesn't guarantee me a sit in the first four rows of the lecture theatre - irony of students.
These got me thinking of how I could get to have a standard note (because ever before I got admission, I was told of how important getting all that is said in a lecture is, making lectures superior to textbooks at times). One day, I thought of building an app that I could install on my phone which would be capable of detecting speech(voice - the words) of my lecturer and translate it into a well-aligned note for me. This still remain something I haven't done because I want something robust and better than any application out there for the same course.
This is what birth my idea - "A full fledged speech Converter into a well aligned note, which would be able to detect things like topic, name of lecturer if this is the first lecture for that course and so on".
Dive into Details
A lot of industry would benefit from this innovation, starting from academics field.
Feature: The Deepgram transcriber
First Use case: Lecture Converter into well-aligned note.
This innovation will help students get standard notes from lectures received. Using the Deepgram's Speech-to-text technology, a lecture received by student can be converted into a note that is free of grammar error.
This will assist students who find it difficult to spell some words to be free; this is one major issue students face in school especially universities because notes are dictated sometimes.
The Deepgram's Speech-to-text technology is expected to be able to detect noise and represent it as - .... or -----, it should also be able to automatically detect when and where punctuation marks should be applied (just like how grammarly detects that).
Second Use Case: Music/songs converter into well-aligned text.
For a while now, I found it difficult to get the lyrics from a song especially rap songs. Having something that works like the shazam app (which returns title of song when played) but for the lyrics of a song, is an amazing innovation Deepgram can proffer.
The Deepgram's Speech-to-text technology will be used to take in speech of song played and automatically converts it to a well-aligned readable text for user. We can never tell this may just sell fast just like the shazam app did. You could go further to make it possible that the Speech-to-text technology could detect and understand any language, because some musicians sing in their native language. It will be cool if Deepgram Speech-to-text technology could detect and understand songs sang in some countries native language into english text which is a universal language.
Just maybe, this might help people like me want to listen to some musicians and this could help their songs go viral, and most importantly let peace continue to grow in the world simply because I understand other people culture just through a song I listened to and loved it.
These ideas will help the music industry, academic industry, the world at large.
Conclusion
The aforementioned ideas can be made and incorporated into API (Application Programming Interface) which could help developers build better apps and one developer might just build the next Shazam for song lyrics, how cool is that. I look forward to getting an opportunity to write something about a feature of the Deepgram Speech-to-text technology, that would be a great privilege.
Thank you Deepgram for making this hackathon happen and if I don't win, am happy I have been able to get an opportunity to drop an idea I have for the Speech-to-text technology. Thanks for this awesome opportunity. In spite of everything, I wish this ideas are brought to life.
God bless Deepgram, keep creating amazing technology.
Top comments (0)