Yesterday, a friend and I built a twitter bot that generates subtitled versions of videos using cloud speech-to-text recognition software. This can be a very helpful tool for the deaf/hard of hearing community, as well as anyone who may not have access to audio on their device. While the bot is far from perfect, we have made reasonable progress and have a few demonstrations on the “media” section of the bot account of strong performance.
On Twitter, I regularly see requests for subtitles in response to videos. Without subtitles, it is virtually impossible for people with hearing disabilities to understand videos published on the internet.
This project will also benefit anyone without access to audio on their device, which can be for a number of reasons.
What I Will Do With $5,000
We need funding in order to afford cloud services for speech-to-text recognition as well as hosting the middleware to connect Twitter’s API to Google Cloud Platform. I expect that $5,000 could keep the service running for at least 5 consecutive years.