I know this is a Pandora's Box but this audience has too much to
offer so here goes....You may have seen the latest project I was on,
the MOTO Q just released by Verizon. I've been working speech and
media audio on Smartphones for a while now. Unlike speech there seems
to be no industry recognized MOS or objective testing tools to validate
media file record/playback minimum system performance. The defacto
benchmark product is the iPod. Other than Compression/ Codec scores,
what tests are done on a new iPod type portable players to assure
consumer audio quality. I'm running into testing issues such as Stereo
Bluetooth accessories & OS Embedded audio DSP width/EQ enhancement.
Besides traditional linearity metrics what tests would you add for
Smartphone based audio players?

Disclaimer: These are my opinions, not Motorola's

Thanks
Kevin T

Kevin T wrote:
I know this is a Pandora's Box but this audience has too much to
offer so here goes....You may have seen the latest project I was on,
the MOTO Q just released by Verizon. I've been working speech and
media audio on Smartphones for a while now. Unlike speech there seems
to be no industry recognized MOS or objective testing tools to validate
media file record/playback minimum system performance. The defacto
benchmark product is the iPod. Other than Compression/ Codec scores,
what tests are done on a new iPod type portable players to assure
consumer audio quality. I'm running into testing issues such as Stereo
Bluetooth accessories & OS Embedded audio DSP width/EQ enhancement.
Besides traditional linearity metrics what tests would you add for
Smartphone based audio players?

The problem is that people all have different notion of what is important
and different musical material has different demands.

I remember when there was a market war between MD and DCC, both using
different compression algorithms. The MD sounded clearly better on
rock music with limited dynamic range. The DCC sounded clearly better on
classical music with wider range.

I don't think there is any way around doing listening tests. The problem
is that you have to do them double-blind with trained listeners and with
a variety of program materials. I think it's also important to present
listeners both the uncompressed and compressed sound one after the other
so they have some baseline for what the compression system is doing.

I like to listen to harpsichords and hear if the attack on the notes gets
smeary. I like to listen to orchestral recordings and see if I can pick
out instruments. I like to listen to Glenn Gould's recording of the Goldberg
variations and see if I can hear the humming. I like to listen to Art
Blakey Live at Birdland and see if I can hear the piano strings resonating
with the horns.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Kevin T wrote:
I know this is a Pandora's Box but this audience has too much to
offer so here goes....You may have seen the latest project I was on,
the MOTO Q just released by Verizon. I've been working speech and
media audio on Smartphones for a while now. Unlike speech there seems
to be no industry recognized MOS or objective testing tools to validate
media file record/playback minimum system performance. The defacto
benchmark product is the iPod. Other than Compression/ Codec scores,
what tests are done on a new iPod type portable players to assure
consumer audio quality. I'm running into testing issues such as Stereo
Bluetooth accessories & OS Embedded audio DSP width/EQ enhancement.
Besides traditional linearity metrics what tests would you add for
Smartphone based audio players?

Disclaimer: These are my opinions, not Motorola's

Thanks
Kevin T

Hi Kevin,

check out:

http://www.avisoft-saslab.com/compression.htm

Also, contact me directly at

mkolber (at) motorola (dot) com

Mark