This past weekend, AssemblyAI was a proud sponsor of Sunhacks, an annual student-run hackathon at Arizona State University. We saw a great turnout at the event with close to 300 participants and 60+ projects built in just under 36 hours.
As a sponsor, we offered our Speech-to-Text API as a resource for students to hack with and provided support for our asynchronous transcription and real-time streaming API. We also had the difficult job of choosing the best project that used the AssemblyAI API effectively.
Here’s a highlight of the top three projects built using the AssemblyAI API:
Live airline crew Speech-To-Text transcription and reading meant for the hearing impaired.
What Does This App Do?
Using live-stream transcription, this application allows individuals on a plane to receive live text-based updates from the flight attendants or the captain as they are speaking. You can create a flight number that is registered into a Cloud Firestore Database with the timestamp and text. As the airline employee is speaking, the application will transcribe the audio and save it to the database, allowing anyone with the same flight number to see announcement updates in real time.
Why This App?
What if the pilot was making announcements and someone with a hearing impairment was onboard? This application uses live-stream transcription to solve a crucial accessibility problem for those with hearing impairment.
Other Applications?
Beyond live in-flight announcements, the transcription can be logged as historical information for passengers to catch up on later. Outside of in-flight services, this application can be extended to airport announcements, train announcements, bus announcements, or any kind of live audio system where important information must be delivered to passengers.
2. Speech Box
Take a break from typing with our intuitive Speech-to-Text browser extension!
What Does This App Do?
- This chrome extension allows users to type with their voice and fill any text box on the web. Using node.js, jQuery, and the AssemblyAI real-time transcription API, users with the chrome extension are able to hold the record button that appears in the text box, and then speak to transcribe the audio stream into text.
Why this App?
- Web accessibility is essential for those with injuries or disabilities, it is also generally helpful for everyone. Even when devices have dictation capabilities, it’s not always the easiest to find and enable the feature. This chrome extension is an easy-to-use alternative with just as accurate live transcription.
3. Discord Clip
Discord bot that records a voice channel and can output a WAV file of the past few seconds.
What Does This App Do?
- The Discord bot is made using Java, the Discord API, Fastjson, AssemblyAI, and Heroku. It listens to your Discord voice calls and allows you to create audio recordings as well as generate the transcript of the recording.
Why This App?
- It’s a fun way to asynchronously communicate with colleagues or friends over Discord, as well as solves an accessibility challenge for those with hearing disabilities on communication platforms.
Other Applications?
- This asynchronous way of communicating through sending video clips and the transcription can be leveraged across different communication platforms like Slack, Microsoft Teams, etc.
We are highly impressed with the quality of all the projects and how students used the AssemblyAI API to solve accessibility and communications challenges we face every day. We can’t wait for another round of Sunhacks next year!
