DVD-Software.info

[esophagus]

any experience or advice for mastering for cheap toy speakers?

I am mastering some real bird sounds for a small cheap speaker toy
device.

After hearing untreated files through cheap speaker, some louder, high
pitch fast change sounds were distorting/peaking. So, right now I am
leveling the audio....reducing the dynamic range a little and EQing out
some harsh (4k- 7k) tweet tweets.

I know this is vague. I am just concered about audio over driving
cheap speaker.

eq range of speaker is 90hz - 12khz and it's a .5 watt speaker.
Speaker is about 1.5 - 2.0 inches in diameter.

other comments?

Thanks,

Craig........ka-kaaaaah (that's a text bird call)

[esophagus]

ah, good questions.

this is for mass release.

manufacturers will compress files for storage for on board chip.

as the files all sound different and have different levels of
background noise (ocean, wind, water, static---those crazy birds!!), I
wonder what the result would be of normalizing all the files with same
setting....mmmm...

I have listened to the unit to get an idea how files sound through set
up. I have treated files in hopes of being most accomodating to
speaker set up.

right now I am compressing with a hard limiter and then bring up levels
a little with l1 maximizier. I then tweak out a little of the file's
noise with noise reduction software. Sometimes for narrow harsh bird
twirps I EQ it out a little so it's a softer sound than harsh. I
know..this is all subjective...but keep the comments coming.

thanks.

[Richard Crowley]

"esophagus" <craigdowning@gmail.com> wrote ...

[Mark Robinson]

Hi Craig,

Here is what I would do. I would ask for a sample of the toy and any
development tools that are used to perform the final file compression. You
need a real world setup which includes the acoustical, electrical, and file
compression artifacts to close the loop in your sound development. Without
the proper means to QC your work, how can they expect good results? If
there is no way to obtain the full system, then I guess I would hack a
monitoring setup using a sample of the toy with the speaker. If possible, I
would try to connect to the amplifier stages inside the toy to get the best
representation of the total sound system. What would be missing is in this
case are file compression artifacts, but maybe you could get away with it.
Perhaps you can determine the target sample rate and bit depth used on the
toy. If you do your sound design at 44.1 Khz /16 bit and the target is
22.05 Khz/8bit, then you can convert to this format on the PC before
monitoring. I would expect that you would want to run your files through a
steep (say 3rd order) low pass and high pass set of filters prior to
compression and limiting. Adobe Audition has a nice set of scientific
filters for this purpose. I would avoid any processing after you downsample.
Hope this helps.
Mark


"esophagus" <craigdowning@gmail.com> wrote in message
news:1131672005.068217.147030@o13g2000cwo.googlegroups.com...