Ayup racers, I'm away for a while so there'll be no crew chief changes for a week or two, and I think it's nearly 'done' in its current incarnation. This leads to an obvious question - what's next? I've been asked a few times to add some kind of 'button event' for it. I've no idea if my background app can see directInput (or whatever it is) events like the user pressing a button on a steering wheel, but it should be possible. So I've been thinking about maybe having a 'tell me stuff' button which triggers the app to give whatever it thinks is a useful bit of info at that point in the race. This would take a bit of work, so is it worth investing the time? I suppose I could have multiple buttons, one for each different bit of info. But it's not exactly 'realistic'... This got me looking into the Microsoft speech recognition API and it does look rather interesting. Basically you define a set of phrases it's listening for (which can have choices / parameters, like "how's my [fuel / tyres / engine temperature]"). You also have to tell it what language and what dialect you'll be speaking in. I'm still at the 'toying with ideas' point here, but it would be *very* cool to press a button on your steering wheel then ask a question into your headset mic, getting the appropriate info back from the crew chief. I still need to establish if this is feasible (technically I *think so* but need to check). If it is feasible it's a lot of work. First I'd need to package it properly in an installer which would also fetch the speech recognition libraries. Then I'd need to have a nice way for the user to select their language and dialect (it might be English only, I don't know). Then I'd need to allow the user to select a 'radio' button on their controller (assuming I can 'hear' button presses while R3E is using the controller). Then I'd start with a small speech recognition grammar with a handful of phrases ("what's my gap [in front / behind]", "how's my fuel", etc). With the button pressed the app would use the MS speech recognition API to listen for a phrase, and if it picks one up with reasonable confidence, it would trigger the message. The hard part of all this is inside the speech recognition engine, which is already done. Whether it works reliably enough I've no idea. For this to be worth investing time in, players would also need a microphone of some kind. The reason I've written all this is that I don't want to do down this route if such functionality wouldn't get used much. If it's just an interesting curiosity then I'll have wasted time I could have spent doing something useful with Crew Chief. What do you guys think?