One of the ideas I had for my Computer Science Honors project was to bring speech recognition/feedback to the iPod. The iPod is a great piece of work from an engineering point of view, but from a usability standpoint, it’s not up to par:
- A user can barely hold the iPod with one hand while scrolling through songs (as opposed to the single-click-and-hold idea). The one thing that irritates me the most is that there are so many mobile devices out there that only seem useful to some weird species of alien with detachable thumbs. Here’s a good test: hold your iPod, your cellphone, whatever, in a position that is most comfortable in your hand — not the way in which you would normally use it. Now attempt to use it. Try and press the buttons. Navigate through the menus. With my “smartphone”, I find myself bending my thumb in ways I never thought possible, and trying to click those teeny buttons. Come on. What were you designers thinking?
- Navigating through the menus is analogous to driving through a really complicated neighborhood, trying to find the house you want. Once you reach a dead end, you have to navigate back the way you came until you find something that seems familiar to you. Why should the end-user be the one who takes the plunge into the sea of menus, successively drilling through them until they reach their goal?
- Driving with an iPod is a good way to wreck your car. Even walking with an Ipod can cause a head-on collision with another person. I find it hilarious that a “mobile” mp3 player was created that actually forced the user to stop everything they were doing to find a song they wanted to play.
- Accessibility. This is an obvious one. Imagine a blind person who enjoys music. The most popular music player on the planet cannot be used by them. There are a couple million people in the US who are so paralyzed that they can only move one finger. Imagine seeing how amazing the iPod is and being unable to scroll through your favorite songs.
So there lies some design problems with the iPod. I believe that some of them can be fixed in a unique way through speech recognition and feedback. The form factor problem, well, can be fixed by getting another mp3 player 
The thing that really drove me the most for this project was the accessibility issue. As a brother to a disabled girl, I’ve witnessed for 21 years how hard it is for people with disabilities to use things we take for granted everyday. It angers me that in this day and age, engineers and designers turn away from people with disabilities, claiming that they’re not in their “market segment”.
Speech feedback would be useful for users with blindness, as well as those people who drive cars and listen to music. As they scrolled through the songs, they could hear exactly what song name is highlighted. They could also hear where, contextually, they are in the song: “20 Percent”, “50 Percent”, etc. to aid the fast-forwarding/rewinding problem. An interesting method would consist of being able to “tag” certain categories with snippets of music, instead of reading them out through a monotone, computer-generated voice.
Speech recognition would also be useful for blind users, and could bring in an entirely new interaction model. Wouldn’t it be awesome if you could simply state into your iPod: “Thievery Corporation, Lebanese Blonde” and it would simply look up the song and play it for you instantly? Goodbye animated menus. You wouldn’t have to do any sort of navigation.
The brute-force strategy of the speech recognition idea would simply involve doing a speech-to-text transformation and performing a search on your songs based on that text, and playing the most relevant.
So how did Apple kill my honors project? The next version of the iPod will have speech feedback. So it didn’t exactly kill it, it just implemented half of it. The speech-to-text will probably exist in a much later version, and would probably be centered around a completely guaranteed approach of performing the speech recognition. Even if they did implement half of the idea, I’m extremely happy. It’s great that companies can realize design issues and fix them in successive versions of their products, especially issues regarding accessibility. I’m still interested in the speech-to-text part of it.. but I have some other interesting ideas for my honors project..
