- Messages sorted by: [ date ][ thread ][ subject ][ author ]
- Next message: Alan Arfken: "Re: active noise cancellation"
- Previous message: Mike Williams: "DD Date Entry Problem"
- Maybe in reply to: Don Colton: "active noise cancellation"
- Next in thread: Alan Arfken: "Re: active noise cancellation"
http://www.sjmercury.com/breaking/docs/060262.htm
Rose
Joseph S. Wisniewski wrote:
>
> Hi Patrick.
>
> This doesn't seem off topic to me, but then again, I've been working in
> automotive speech recognition for 8 years.
>
> The problems you outline, sharing attention between the driving task and
> running the speech system, are fairly well documented. It can be a bit
> tricky. That's why we usually confine automotive voice applications to
> areas where operating a control without a voice system takes the eyes,
> hands, and mind away from driving already. Cellular telephone is a
> perfect example. Some of the more complicated climate control systems and
> stereo systems are other good candidates. Dictation, while doable,
> requires some very careful attention to human factors to make a uesable
> system. Correction can really be a major pain.
>
> Suprisingly, the automotive noise environment isn't that bad when you
> have a headset mic availiable. noise levels (except for fairly high
> speeds) are about office level. Unfortunatly, wearing a headset while
> driving is not legal in much of the world (some places even outlaw
> headsets that don't cover either ear, just provide a place for the boom
> mike). Even if it's legal, it isn't acceptable to many drivers, and not a
> fun thing to do over the entire automotive temperature range (headset at
> -40 degrees, anyone?)
>
> Things get much uglier when you try to place a microphone at some
> distance from the driver's mouth. By the time you finish juggling the
> safety, styling, manugacturability, and reliability concerns, the mic
> ends up some 20 inches (50cm) from the driver's mouth. That takes a good
> mic, and a good recognition front end.
>
> On Sat, 21 Feb 1998, Patrick Maheral wrote:
>
> > This might be a little off topic, but I tried dictating (into an
> > MiniDisk recorder) while driving (is this considered an impairment?).
> > I would not recommend this. I found it difficult to concentrate on
> > driving and on what I was dictating (eg. punctuation, spelling proper
> > names, etc). At least the first 5 minutes, I found that I was
> > thinking more about dictation than the light traffic. After
> > relaxing, my driving improved (I hope).
> >
> > Anyway, I played the MD recording back to NatSpeak Deluxe (I have a
> > user trained as MD recordings) and found the accuracy to be OK. I
> > was using a VXI mic which claims to be a noise cancelling mic
> > (though not active I think). I thought there would be a lot of
> > background noice such as wind, engine, and road noise, and the radio.
> > To my surprise, I couldn't hear any of these on the recording. I
> > guess the (passive?) noise cancelling did its job.
>
> Joseph S. Wisniewski | Views expressed are my own, and don't reflect
> Ford Motor Company | those of the Ford Motor Co. or affiliates.
> Project Sapphire | LeMans, Daytona, Bonneville, and Sebring are
> jwisniew@@ford.com | just races, won by people driving Ford cars!
- Next message: Alan Arfken: "Re: active noise cancellation"
- Previous message: Mike Williams: "DD Date Entry Problem"
- Maybe in reply to: Don Colton: "active noise cancellation"
- Next in thread: Alan Arfken: "Re: active noise cancellation"
 |  |