[prev in list] [next in list] [prev in thread] [next in thread] 

List:       festlang-talk
Subject:    [festival-talk] pitchmarks + lpc
From:       nshmyrev () yandex ! ru (Nickolay V !  Shmyrev)
Date:       2006-03-15 22:22:26
Message-ID: 1142461346.24180.15.camel () localhost ! localdomain
[Download RAW message or body]

message from "Nickolay V. Shmyrev" <nshmyrev at yandex.ru> to festival-talk
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
? ???, 15/03/2006 ? 20:33 +0100, Stefan Pal ?????:
> message from "Stefan Pal" <stpa01ad at student.cbs.dk> to festival-talk
> = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
> Hi,
> 
> I recorded a voice with a sampling rate of 44Khz. should I modify the order
> in the make_lpc script which by default is 16?
> 
> for i in $*
> do
>    fname=`basename $i .wav`
>    echo $i
> 
>    # Potential normalise the power
>    #$ESTDIR/bin/ch_wave -scaleN 0.5 $i -o /tmp/tmp$$.wav
>    # resampling can be done now too
>    #$ESTDIR/bin/ch_wave -F 11025 $i -o /tmp/tmp$$.wav
>    # Or use as is
>    cp -p $i /tmp/tmp$$.wav
>    $ESTDIR/bin/sig2fv /tmp/tmp$$.wav -o lpc/$fname.lpc \
>              -otype est -lpc_order 16 -coefs "lpc" \ 
>              -pm pm/$fname.pm -preemph 0.95 -factor 3 \
>              -window_type hamming
>    $ESTDIR/bin/sigfilter /tmp/tmp$$.wav -o lpc/$fname.res \
>               -otype nist -lpcfilter lpc/$fname.lpc -inv_filter
>    rm /tmp/tmp$$.wav
> done
> 

You can see that resompling can be done in this script also, this line
does exactly that 

   #$ESTDIR/bin/ch_wave -F 11025 $i -o /tmp/tmp$$.wav

>From my experience, it's possible to create voice from 44 kHz and it will be a bit
better than 16 kHz version, but I may be wrong Alan will correct me. Generally you 
can just try different values and see what is better for you. 44 kHz version will
be bigger certainly without much quality improvement. 


> Another question is pitchmarks. I have read the documentation, but I'm a bit
> confused about the values in the pitchmark-script.
> 
> -min 0.005 -max 0.012 -fill -def 0.01 \
> -wave_end -lx_lf 200 -lx_lo 71 -lx_hf 80 -lx_ho 71 -med_o 0
> 
> Are the values in Hz? And how do I set the min and max correctly?
> 

That is the nature of festival - everything is just magic, you should
try different things and learn. Some words about it are located at

http://festvox.org/bsv/bsv-pitchmarks-sect.html

the lx_lf and lx_hf are filter frequences, they are in Hz, lx_lo and
lx_ho are filter order, they are just naturals :), btw pitchmark command
that do actual work have --help argument that explains the same.

> Thanks in advance,
> 
> Stefan Pal 
> 
> = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
> =    University of Edinburgh's Festival Speech Synthesis System       =
> = http://festvox.org/festival      Sent Via festival-talk at festvox.org =
> =                           To unsubscribe mail majordomo at festvox.org =
> = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
> 

= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
=    University of Edinburgh's Festival Speech Synthesis System       =
= http://festvox.org/festival      Sent Via festival-talk at festvox.org =
=                           To unsubscribe mail majordomo at festvox.org =
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =



[prev in list] [next in list] [prev in thread] [next in thread]