Comments on: Voice Code http://uiscape.com/2007/05/02/voice-code/ Research notes on interaction design Mon, 06 Oct 2008 22:46:22 +0000 http://wordpress.org/?v=2.1 By: pauldwaite http://uiscape.com/2007/05/02/voice-code/#comment-38 pauldwaite Mon, 07 May 2007 10:31:45 +0000 http://uiscape.com/2007/05/02/voice-code/#comment-38 Is there any chance this would be faster than typing? If not, I'd suggest it's got zero chance of any significant use. Is there any chance this would be faster than typing? If not, I’d suggest it’s got zero chance of any significant use.

]]> By: Amar http://uiscape.com/2007/05/02/voice-code/#comment-39 Amar Mon, 07 May 2007 13:38:57 +0000 http://uiscape.com/2007/05/02/voice-code/#comment-39 I think their main target audience seem to be programmers who suffer from RSI. I can also see this getting used by people with physical disabilities who cannot use a mouse or keyboard. There's <a href="http://dx.doi.org/10.1145/1168987.1169022" title="Hubbell, Langan & Hain (2006). A voice-activated syntax-directed editor for manually disabled programmers" rel="nofollow">another system</a> that targets these users specifically. Still, it would indeed be interesting to know whether it's fast enough to tempt even "manual" programmers. Unfortunately, their paper doesn't talk about this. I think their main target audience seem to be programmers who suffer from RSI. I can also see this getting used by people with physical disabilities who cannot use a mouse or keyboard. There’s another system that targets these users specifically.
Still, it would indeed be interesting to know whether it’s fast enough to tempt even “manual” programmers. Unfortunately, their paper doesn’t talk about this.

]]> By: Sven Ryen http://uiscape.com/2007/05/02/voice-code/#comment-40 Sven Ryen Wed, 09 May 2007 20:34:31 +0000 http://uiscape.com/2007/05/02/voice-code/#comment-40 On Phones runnnig Nokia S40 Series 3rd Edition, the voice recognition is quite good. The correct name is usually picked, unless you have many similar names in your address book, like "Paul Johnsen" and "Paul Johnsted". In that case, the programmers have been careful enough to introduce a small delay in which the user can select the correct contact from the phone menu - if he happens to be looking at his screen. it's to my knowledge not possible to "cancel" an outgoing call by voice commands. The headset will duly repeat the phrase the phone believes to have picked up, and then - if it has picked the wrong contact - you'll have to immediately grab for your phone to hang up the call before the wrong person answers. On Phones runnnig Nokia S40 Series 3rd Edition, the voice recognition is quite good. The correct name is usually picked, unless you have many similar names in your address book, like “Paul Johnsen” and “Paul Johnsted”.
In that case, the programmers have been careful enough to introduce a small delay in which the user can select the correct contact from the phone menu - if he happens to be looking at his screen.
it’s to my knowledge not possible to “cancel” an outgoing call by voice commands. The headset will duly repeat the phrase the phone believes to have picked up, and then - if it has picked the wrong contact - you’ll have to immediately grab for your phone to hang up the call before the wrong person answers.

]]> By: Alexander http://uiscape.com/2007/05/02/voice-code/#comment-42 Alexander Wed, 09 May 2007 22:06:53 +0000 http://uiscape.com/2007/05/02/voice-code/#comment-42 Paul: I don't have any hard data at hand (although I guess they do exist) about programmer productivity, but from my personal experience writing software, I doubt that the #1 goal is the speed of data entry. Modern IDEs (such as Eclipse) have already reduced the need for a lot of typing by providing code templates and context-sensitive completions, features also present in VoiceCode. I don't think that triggering these features by voice is going to be much slower than triggering them with a couple of keystrokes. Sven: Good observation, as I said, there always needs to be a quick way of correcting mistakes made by voice recognition. Although I suspect that if you're talking to the phone through a headset, all headsets should have a button that allows you to end the call. Paul: I don’t have any hard data at hand (although I guess they do exist) about programmer productivity, but from my personal experience writing software, I doubt that the #1 goal is the speed of data entry. Modern IDEs (such as Eclipse) have already reduced the need for a lot of typing by providing code templates and context-sensitive completions, features also present in VoiceCode. I don’t think that triggering these features by voice is going to be much slower than triggering them with a couple of keystrokes.

Sven: Good observation, as I said, there always needs to be a quick way of correcting mistakes made by voice recognition. Although I suspect that if you’re talking to the phone through a headset, all headsets should have a button that allows you to end the call.

]]> By: VoiceCode: Advances in Voice Recognition Software at The Ergonomenon http://uiscape.com/2007/05/02/voice-code/#comment-49 VoiceCode: Advances in Voice Recognition Software at The Ergonomenon Sat, 19 May 2007 00:38:01 +0000 http://uiscape.com/2007/05/02/voice-code/#comment-49 [...] UIScape: Researchers from the National Research Council of Canada have created VoiceCode, an application [...] […] UIScape: Researchers from the National Research Council of Canada have created VoiceCode, an application […]

]]>