<html dir="ltr"><head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<meta name="GENERATOR" content="MSHTML 8.00.7601.17720">
<style title="owaParaStyle"><!--P {
MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
--></style>
</head>
<body ocsi="x">
<div dir="ltr"><font color="#000000" size="2" face="Tahoma">Wow - nice! A few quick questions:</font></div>
<div dir="ltr"><font size="2" face="tahoma"></font> </div>
<div dir="ltr"><font size="2" face="tahoma">1. How long can the recording be for translation?
</font></div>
<div dir="ltr"><font size="2" face="tahoma">2. Any limitation on how much text the return (transcribed) variable can hold?</font></div>
<div dir="ltr"><font size="2" face="tahoma">3. Any commercial / terms of use limitations?</font></div>
<div style="DIRECTION: ltr" id="divRpF731611">
<hr tabindex="-1">
<font size="2" face="Tahoma"><b>From:</b> asterisk-users-bounces@lists.digium.com [asterisk-users-bounces@lists.digium.com] On Behalf Of Bruce B [bruceb444@gmail.com]<br>
<b>Sent:</b> Wednesday, January 04, 2012 1:25 PM<br>
<b>To:</b> Asterisk Users List<br>
<b>Subject:</b> Re: [asterisk-users] Speech recognition in asterisk using google voice API<br>
</font><br>
</div>
<div></div>
<div>
<div class="gmail_quote">
<blockquote style="BORDER-LEFT: #ccc 1px solid; MARGIN: 0px 0px 0px 0.8ex; PADDING-LEFT: 1ex" class="gmail_quote">
Note to self: "Never release anything asterisk related without testing<br>
on RHEL/Centos 5"<br>
<br>
Thank you for reporting this. I have replaced sox with flac and it seems<br>
to work now on older platforms too (tested on Centos 5 with asterisk 1.4).<br>
You can get the updated code here:<br>
<a href="https://github.com/zaf/asterisk-speech-recog/tarball/master" target="_blank">https://github.com/zaf/asterisk-speech-recog/tarball/master</a><br>
<div>
<div></div>
<div><br>
----------------<br>
Lefteris Zafiris<br>
</div>
</div>
<div><br>
</div>
</blockquote>
<div><br>
</div>
<div>Works beautifully. Amazing job Lefteris. Thanks.</div>
<div><br>
</div>
<div>The best result I got in probability was 0.9725632 by saying, "hello". I think there is some non-phonetic logic built-in as well. I tried, "1, 2" and I got "0.86534226" in accuracy. While I tried "1, 2, 3, 4, 5" I got, "0.97256315". Probably Google sees
the pattern?!</div>
<div><br>
</div>
<div>What are some of the other tricks (if any) or consideration that one should make while creating a strong speech recognition enabled IVR?</div>
<div><br>
</div>
<div>Best,</div>
</div>
<br>
</div>
</body>
</html>