(keitai-l) Re: Sound files on FOMA handsets

hi,

various tools to embed wav data in an MFi file 
and instructions how to use them can be found at

http://kamen.dyndns.org/voice/index.html

it's a japanese site but easy to understand when web
translated.

data are uncompressed so it's simple to calculate the
possible sound lenght:
the usual format is 16 bit, 8 khz, mono which means that
one second would take 16,000 bytes, plus some 100 bytes
for the MFi header.
so within an i-appli jar file there would be space for
maximum 0.6 seconds of total sound length, depending on
the code size.
this value can be doubled by reducing sampling frequency
to 4 khz but sound is already awful at 8 khz..

however, one could load the sound samples from a server,
use a separate AudioPresenter for each spoken word
and play them after each other.
i'm not sure but i think there's no 10 kb limit for sound
files when loaded from a server, maybe you could even fit
a whole sentence into one file this way.
this would avoid memory problems but users would have to
download sound_length_in_seconds * 16 kilobytes each
time they use the i-appli.

for those keitai with a rohm soundchip (panasonic &
mitsubishi, not sure about their FOMA models) there's no
way to play audio data in i-appli, they only have a fixed
preset of wavetable midi sounds.


cheers

oliver wittchow



> Subject: Re: Sound files on FOMA handsets
> Date: Sat, 2 Feb 2002 15:42:01 +0900
> 
> Some suggested search engine terms:
> 
>  ADPCM
>  MFi
>  着メロ
> 
> Will this get you very far?  I haven't really
> tried, so I don't know.  It does appear that
> you can put something like sound-waveform
> data into melody files.  I've been told, however,
> that this yields only a few seconds of low
> quality output.
> 
> This state of affairs is almost certainly
> intentional.  It's a bit maddening, but phone
> companies actually want to be able to charge
> you for voice communications of almost
> any kind.
> 
> -michael turner
> leap@gol.com