I once experimented at the Mail.Ru Tech Forum with transcribing speeches into text near real-time. Several operators were working in parallel on pieces of video, which were then assembled together.
Now I’m watching CNN, and they have subtitles running in real time. How do they do it? I went online to find out, and it turns out that stenotypes – keyboards + software for translating live speech into text – have been around for fifty years. And it’s managed by just one operator.
The main use of stenotypes is in courts. Here’s a link – a very interesting video about how to type text on this stenotype. I’m sure what you imagine doesn’t match how it actually is.
Hi, I'm Rauf Aliev. 