it works with gi=n.cumsum(dD,0,<a href="http://n.int" target="_blank">n.int</a>)%Dv<br>for me.<br><br>it was not working beforehand, in this version:<br><a href="https://gist.github.com/3815747/9721cfc5eff7a4dd706a7f1e80a6264d3bd6915d">https://gist.github.com/3815747/9721cfc5eff7a4dd706a7f1e80a6264d3bd6915d</a><br>
when i assumed frequency variation and did the LUT<br>procedure normally, without the cumsum which takes in<br>account individual steps and resulting positions.<br><br>can you post the .wav that <a href="https://gist.github.com/3815747">https://gist.github.com/3815747</a><br>
outputs?<br><br>thanks for the feedback and the C version!<br><br><div class="gmail_quote">2012/10/2 SxDx <span dir="ltr">&lt;<a href="mailto:sed@free.fr" target="_blank">sed@free.fr</a>&gt;</span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
The last one in the series, sorry for the mess...<br>
<br>
so replace:<br>
gi=n.cumsum(dD,0,<a href="http://n.int" target="_blank">n.int</a>)%Dv<br>
by:<br>
gi=n.cumsum(dD)<br>
gi=n.array(gi, dtype=int)%Dv<br>
</blockquote></div><br><br clear="all"><br>-- <br>GNU/Linux User #479299<br><a href="http://labmacambira.sf.net" target="_blank">labmacambira.sf.net</a><br>