Re: AUDITORY Digest - 28 Nov 2001 to 29 Nov 2001 (#2001-211)

[Jason Summers <summej2@xxxxxxx>]

>
> Dear List,
>
> We are conducting research on auditory model based front-ends for
> automatic speech recognition in reverberant acoustic conditions.
...
> Thanks in advance!
> Regards,
> Kalle Palomaki
> ---
Kalle, I'm not so sure about the signal to noise ratio of the impulses which
you found on the web.  For convolution with signals we like to have S/N >=
60dB at mid-frequencies.  These we will be fairly difficult to find.  Yamaha
has produced a set of IR's for use with their new convolver that are quite
nice but I don't believe that they are available any other way than
purchasing the unit.

 I may personally have some of the IR's you would need from some of my own
work.  What sort of source to receiver distances are you looking for?  All
of the possible IR's would have source to receiver distances that are quite
large (performer to audience or lecturer to student for example).

Jason Summers
Rensselaer Polytechnic Institute
>
> Subject: Measuring Reverberation time
>
> I am looking for references for appropriate techniques for measuring =
> reverberation time of a medium sized room (approx. 40'x40'x20'). In =
> specific, number and position of loudspeakers, level and type of signal. =
> Does your technique and stimuli vary depending on your question of =
> interest (e.g. absorption characteristics versus effects on sound =
> quality or speech intelligibility)? We are interested in speech =
> intelligibility by the way. I have ordered an IEC standard that =
> supposedly describes a "standard" method for measuring reverb times when =
> examining the absorption characteristics of materials.  Any thoughts are =
> appreciated.
>
> Thanks,
>
> Ben Hornsby

Ben,

These days most of us investigating room acoustics measure the impulse
response of a room and derive measures like RT from the impulse response.
Measuring the IR correctly will require some specialized equipment however:
omnidirectional loudspeaker, etc. while it would be possible to get an
approximation of RT much more simply.  It is a question of desired precision
and accuracy.

ISO 3382 is the best standard I know of for measurement of both
reverberation time and impulse response.  However, the latest edition of
Kuttruff's Room Acoustics will have much of the information in the ISO
standard and far more background information.  The number of measurement
positions will depend both on the size of the room since RT varies
statistically through a space.  Both Kuttruff's book and the ISO standard
detail this.

Jason Summers
Rensselaer Polytechnic Institute