Speech Recognition Robots
One of the blogs I read and enjoy is Tod Maffin's "I Love Radio .org"; he's a freelance radio journalist / producer, with an interest in podcasts and podcasting, and in my opinion posts some really entertaining content. He has a post entitled "Talking to Speech Recognition Robots", which includes a downloadable 5 minute MP3 podcast containing an episode in which he tries (unsuccessfully!) to communicate with such a system. Quite apart from the amusement factor at his evident frustration, this particular piece interests me on two fronts; one professional, the other strictly personal.
On the first front, many of Exony's customers use some form of automated call handling system; for example, based on touch-tone recognition, speech recognition, or Caller Line Identification (CLI). Some use the kind of speech recogniion robots Tod wails about. The business case for using these technologies can be overwhelming, from the perspective of the *organisation*; but how often do the poor users who have to do battle with these systems feel less than enthused by their experience? It really *is* in the organisation's best interest to deploy such technologies in a way that users can (and will) use; because the alternative (where users typically do something erroneous to try to make the automated system connect them to a human) is typically not what he organisation wants. Neither, really, is it in the user's best interests, because the likelihood is, they could transact their business more quickly and effectively if only the automated systems were clear, reliable and appropriate. (Disclaimer: there are probably as many examples of organisations which deploy such technology with great success as there are instances of heroic failures!)
On the second front, I hate using the telephone these days. I always did, primarily because I am a) an auditory communicator (which means that the primary sense through which I experience, process and respond to the world is through hearing and sound) and b) a "high- or no-bandwidth communicator" (which means I either take in information, content and events very quickly, or periodically can't take them in at all). When I'm in high-bandwidth mode, most people seem to use too many words to say too little and (without other stimuli such as body language to augment the conversation) I get very frustrated, very quickly. In no-bandwidth mode (basically, when I am in or recovering from one of my occasional but quite debilitating episodes of Depression) I can hardly process at all; certainly not speak in response. Hence, the telephone has always been a bit of a dead loss for me. (I know, women are supposed to love chatting on the phone; not this one.)
Currently, the phone represents a real problem for me. My speaking voice is, naturally, quite deep. I will be undergoing professional speech therapy to correct this, as well as to learn more female patterns of speech and inflection. Until then, my voice is a bit of a giveaway; the telephone seems to exaggerate this to the n'th degree. Rarely, if ever, do I "pass" on the phone. Ordinarily, this wouldn't be a problem (I'm not ashamed of who or what I am, and anyway this is a transitory phase) but I encounter problems with organisations I have to do personal business with on an almost daily basis. I speak to them (telephone company, bank, insurance company, roadside assistance organisation in recent weeks) and we get to the part where they have my details on screen, and then comes the pause; then, "what relationship are you to the account / policy holder?" "I *am* the account / policy holder". "But it says 'Miss' here..." "Yes, that's correct." "But you're a... you're a..." at which point I either pedantically stand my ground, or else I have to explain my personal situation... again. Put a note on their records so the next time I have to call them they can see an explanation? Do me a favour. The worst part of this is that an excuse sometimes given by the companies for this behaviour on the part of their employees is to ensure data protection; *my* data, let's not forget. Excuse me, but maybe there's some data I'd like to keep private, too; like, not having to explain out loud, maybe in a public place, the reason for my apparently discordant voice. It's more than merely annoying; it could put me in personal danger. Wake up, people, this is 2004, I don't have two heads, I'm not a criminal and I'm not a threat to national security.
From which rant you might gather that I, for one, would *love* to deal with (reliable) speech recognition robots when I have to use the telephone! :-)
This blog has been migrated to new software on a different server (http://www.multidimensional.me.uk) and comments on this post on *this* blog are now closed. All existing comments have been copied to the equivalent post on the new blog. If you still wish to comment on this post, please use the equivalent post at: http://www.multidimensional.me.uk/
On the first front, many of Exony's customers use some form of automated call handling system; for example, based on touch-tone recognition, speech recognition, or Caller Line Identification (CLI). Some use the kind of speech recogniion robots Tod wails about. The business case for using these technologies can be overwhelming, from the perspective of the *organisation*; but how often do the poor users who have to do battle with these systems feel less than enthused by their experience? It really *is* in the organisation's best interest to deploy such technologies in a way that users can (and will) use; because the alternative (where users typically do something erroneous to try to make the automated system connect them to a human) is typically not what he organisation wants. Neither, really, is it in the user's best interests, because the likelihood is, they could transact their business more quickly and effectively if only the automated systems were clear, reliable and appropriate. (Disclaimer: there are probably as many examples of organisations which deploy such technology with great success as there are instances of heroic failures!)
On the second front, I hate using the telephone these days. I always did, primarily because I am a) an auditory communicator (which means that the primary sense through which I experience, process and respond to the world is through hearing and sound) and b) a "high- or no-bandwidth communicator" (which means I either take in information, content and events very quickly, or periodically can't take them in at all). When I'm in high-bandwidth mode, most people seem to use too many words to say too little and (without other stimuli such as body language to augment the conversation) I get very frustrated, very quickly. In no-bandwidth mode (basically, when I am in or recovering from one of my occasional but quite debilitating episodes of Depression) I can hardly process at all; certainly not speak in response. Hence, the telephone has always been a bit of a dead loss for me. (I know, women are supposed to love chatting on the phone; not this one.)
Currently, the phone represents a real problem for me. My speaking voice is, naturally, quite deep. I will be undergoing professional speech therapy to correct this, as well as to learn more female patterns of speech and inflection. Until then, my voice is a bit of a giveaway; the telephone seems to exaggerate this to the n'th degree. Rarely, if ever, do I "pass" on the phone. Ordinarily, this wouldn't be a problem (I'm not ashamed of who or what I am, and anyway this is a transitory phase) but I encounter problems with organisations I have to do personal business with on an almost daily basis. I speak to them (telephone company, bank, insurance company, roadside assistance organisation in recent weeks) and we get to the part where they have my details on screen, and then comes the pause; then, "what relationship are you to the account / policy holder?" "I *am* the account / policy holder". "But it says 'Miss' here..." "Yes, that's correct." "But you're a... you're a..." at which point I either pedantically stand my ground, or else I have to explain my personal situation... again. Put a note on their records so the next time I have to call them they can see an explanation? Do me a favour. The worst part of this is that an excuse sometimes given by the companies for this behaviour on the part of their employees is to ensure data protection; *my* data, let's not forget. Excuse me, but maybe there's some data I'd like to keep private, too; like, not having to explain out loud, maybe in a public place, the reason for my apparently discordant voice. It's more than merely annoying; it could put me in personal danger. Wake up, people, this is 2004, I don't have two heads, I'm not a criminal and I'm not a threat to national security.
From which rant you might gather that I, for one, would *love* to deal with (reliable) speech recognition robots when I have to use the telephone! :-)
This blog has been migrated to new software on a different server (http://www.multidimensional.me.uk) and comments on this post on *this* blog are now closed. All existing comments have been copied to the equivalent post on the new blog. If you still wish to comment on this post, please use the equivalent post at: http://www.multidimensional.me.uk/
posted by Koan @ 7:55:00 am
0 Comments:
<< Home