Converting spoken language to visible sound waves to aid deaf

When watching TV, I sometimes enable subtitles for hearing impaired (not because i suffer from any hearing loss, but sometimes i like reading better than hearing).

When this happens to be a live broadcast, subtitles are often not available, or have a (understandable) noticeable lag when they are being produced on the fly. Not a problem for me, i can simply turn on the volume.

I've been thinking, though, how helpful it would be if there was a live, visible representation of the audio. A simple visible sound wave is probably way too simplistic to make out words and phrases, but if it were possible to more distinctly visualize sounds, my idea is that it might be possible to learn to read this visual output, and thereby being able to follow more directly whatever is spoken.

Does something like that exist?