Through The Daily WTF, I have become aware of Hear A Blog, a service that produces human narration of written content for people to listen to. I am sure there are many people for whom this is useful: the blind, people who want the book-on-tape paradigm to apply to timely web content, etc. Personally, I desperately want someone to develop a Read A Podcast service. (I also hate how Apple managed to brand the old and obvious idea of expressing non-musical information in recorded audio form, but that is a fight for another day.)
There are all sorts of audio streams that provide information I would find interesting. On the site Slate alone there are several weekly conversations that look interesting to me. Every week they have a discussion of sports stories from that week, a discussion of cultural and media stories, one on politics, and another about food. A lot of information would be lost in a transcript, but I can live without inflection, phrasing, and the like. Another rich source of non-textual information is in video. One of the blogs that I read, when it updates, very often points out clips on youtube or TED talks or such things where there may be interesting visuals but the essential content is just someone talking.
I am glad that audio and video content exists for those people who can best process information provided to them in an auditory or visual form, but I feel like text should be the universal lowest common denominator in which all information is available. There is simply no way that I am going to take half an hour out of my day to watch a video of someone talking or listen to a conversation that does not involve me. If I could instead spend five minutes reading what had been said, however, that would be very attractive. For major media outlets like Slate, producing a transcript would be a very minor cost. For content repositories like youtube it would be very easy to run a speech recognition algorithm on all of their data and provide estimated transcriptions of those videos that contain at least 25% speech. If I had more time and ambition, I might try to build a crowd-sourced community of people that transcribe whatever they read / watch for the benefit of others so that only one member of the community needs to go through that per piece of media.