<html><head><style type="text/css"><!-- DIV {margin:0px;} --></style></head><body><div style="font-family:times new roman,new york,times,serif;font-size:12pt">Hi Wellie,<br><br>Many thanks for purchasing Vestec speech recognition engine.<br><br>Respectfully, I would like to make a few clarifications: <br><br>(a) Vestec speech engine is designed for IVR-type interactive voice applications. Almost all such applications can be easily addressed with a vocabulary size of 500 keywords per recognition. (Remember, there can be several recognitions within one application; each recognition pertains to one state, such as one level in a multi-level menu tree). In addition, a variety of non-IVR applications - such as speech-based directory assistance (ie. utterance of a person's name as opposed to entering the customary first three letters of a person's last name for routing) - can be built for all SMB and most SME applications with a speech engine supporting 500
keywords vocabulary size.<br><br>(b) Vestec speech engine is not designed to be used as a general purpose dictation/transcription system. Such dictation/transcription systems - for example, for voicemail-to-email transcription - generally require a vocabulary size of tens of thousands of words, not to mention important "training" data (that must be collected or purchased separately) for building the grammar for the desired application. Nuance offers a dictation system in the form of its its Dragon Speech product.<br><br>(c) in the IVR space - which by the way constitutes the overwhelming majority of commercial speech applications and is a natural fit with telephony platforms such as Freeswitch - Vestec speech engine offers the best deal around. Not only does it offer among the highest recognition accuracy in the industry and industry standards-based grammar writing format, it supports the same vocabulary size as LumenVox Lite (500 keywords) but costs
less than 50% of its retail price, and offers twice the vocabulary size of Nuance tier-1 engine (250 keywords) and retails at less than 20% of its price.<br><br>Hope this helps.<br><br>Best regards,<br>-Kashif<br><div> </div>Kashif Kahn<br>VP, Business Development<br>Vestec, Inc.<br>Waterloo, ON Canada<br>phone: (519) 885-7615<div><br></div><div style="font-family: times new roman,new york,times,serif; font-size: 12pt;"><br><div style="font-family: arial,helvetica,sans-serif; font-size: 13px;"><font face="Tahoma" size="2"><hr size="1"><b><span style="font-weight: bold;">From:</span></b> Wellie Chao <wchao@yahoo.com><br><b><span style="font-weight: bold;">To:</span></b> freeswitch-users@lists.freeswitch.org<br><b><span style="font-weight: bold;">Sent:</span></b> Tue, May 11, 2010 9:03:05 PM<br><b><span style="font-weight: bold;">Subject:</span></b> Re: [Freeswitch-users] Voice to text / speech recognition<br></font><br>Thanks! I am installing
Vestec's solution now. It only handles small <br>vocabulary speech recognition rather than large vocabulary speech <br>recognition, but that's also useful (albeit for a different service). We <br>can use it for IVR menus. Still looking for a large vocabulary solution <br>similar to Nuance's product. I'd love to hear from anybody on the list who <br>has used a large vocabulary speech recognition product.<br><br><br><br>Date: Mon, 10 May 2010 16:46:12 -0700<br>From: Michael Collins <<a ymailto="mailto:msc@freeswitch.org" href="mailto:msc@freeswitch.org">msc@freeswitch.org</a>><br>Reply-To: <a ymailto="mailto:freeswitch-users@lists.freeswitch.org" href="mailto:freeswitch-users@lists.freeswitch.org">freeswitch-users@lists.freeswitch.org</a><br>To: <a ymailto="mailto:freeswitch-users@lists.freeswitch.org" href="mailto:freeswitch-users@lists.freeswitch.org">freeswitch-users@lists.freeswitch.org</a><br>Subject: Re: [Freeswitch-users] Voice to text /
speech recognition<br><br>You guys need to check out Vestec. Go buy a $25 SDK for your dev server. Tell them that the FS guys sent you and they'll refund<br>your $25 (check via snail mail). I haven't had a chance to dig into this one yet but I fully intend to do so. However, if you<br>guys are working on real solutions to real world problems then you're in a great position to give Vestec a test drive. More<br>info: <a href="http://www.freeswitch.org/node/252" target="_blank">http://www.freeswitch.org/node/252</a><br><br>See, you should visit freeswitch.org more frequently than twice a year! ;)<br><br>-MC<br><br>On Mon, May 10, 2010 at 3:49 PM, Jan Berger <<a ymailto="mailto:jan.berger@video24.no" href="mailto:jan.berger@video24.no">jan.berger@video24.no</a>> wrote:<br> Hi,<br><br> I am going to look into Sphinx myself for IVR menus.<br><br> Recognition factor is not so much the software
as it is the quality of the<br> database and how well it is tuned. Tuning you can do yourself with tools<br> provided from sphinx + it does exist a separate open source database that I<br> believe you can use.<br><br> Doing IVR menus is feasible. But, to translate free speech you have a<br> challenge called dialects. The fact is that recognition factor is not 100%<br> even with people with the same dialect, and many dialects are very far out<br> so analyze the language of your clients and make some very accurate<br> questions to the vendor to ensure you get what you want.<br><br> Jan<br><br>-----Original Message-----<br>From: <a ymailto="mailto:freeswitch-users-bounces@lists.freeswitch.org"
href="mailto:freeswitch-users-bounces@lists.freeswitch.org">freeswitch-users-bounces@lists.freeswitch.org</a><br>[mailto:<a ymailto="mailto:freeswitch-users-bounces@lists.freeswitch.org" href="mailto:freeswitch-users-bounces@lists.freeswitch.org">freeswitch-users-bounces@lists.freeswitch.org</a>] On Behalf Of Wellie<br>Chao<br>Sent: 10. mai 2010 19:09<br>To: <a ymailto="mailto:freeswitch-users@lists.freeswitch.org" href="mailto:freeswitch-users@lists.freeswitch.org">freeswitch-users@lists.freeswitch.org</a><br>Subject: [Freeswitch-users] Voice to text / speech recognition<br><br>I'm looking to provide a feature to transcribe voicemails into text for my<br>users. I've looked at SpinVox (which was acquired by Nuance). Their<br>licensing method doesn't seem particularly friendly or flexible. They<br>charge per user and lock each license to a particular user's actual<br>identity. If I have a user named John Smith, I have to buy a license from<br>Nuance _for
John Smith_. If John Smith stops using the voicemail to text<br>feature, but I have another user named Jane Doe who wants to start using<br>the voicemail to text feature, I have to buy a new license for Jane Doe.<br>Since Nuance makes you pay for a license one year at a time, it gets<br>expensive if you have any significant churn. Apart from that, it is just<br>annoying to have to register users with another company. I'd like to buy<br>100 or 1000 or 10000 licenses and use them how I see fit and not have to<br>bother with registering individual users with a provider of speech<br>recognition services.<br><br>I am wondering if other FreeSWITCH users have recommendations for good<br>voice to text services or software. I'd be happy to consider either a<br>service or software. If software, ideally it would be free of course, but<br>a reasonable cost would also be acceptable. I know about PocketSphinx, but<br>is it really sufficiently high quality as to be
useful for transcribing<br>voicemails? I don't need 100% accuracy -- probably 80% or 90% would be<br>good enough since I will also attach the WAV file and direct my users to<br>use the WAV file when in doubt.<br><br>Any pointers or tips would be appreciated.<br><br>_______________________________________________<br>FreeSWITCH-users mailing list<br><a ymailto="mailto:FreeSWITCH-users@lists.freeswitch.org" href="mailto:FreeSWITCH-users@lists.freeswitch.org">FreeSWITCH-users@lists.freeswitch.org</a><br><a href="http://lists.freeswitch.org/mailman/listinfo/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/listinfo/freeswitch-users</a><br>UNSUBSCRIBE:<a href="http://lists.freeswitch.org/mailman/options/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/options/freeswitch-users</a><br><a href="http://www.freeswitch.org"
target="_blank">http://www.freeswitch.org</a><br><br><br>_______________________________________________<br>FreeSWITCH-users mailing list<br><a ymailto="mailto:FreeSWITCH-users@lists.freeswitch.org" href="mailto:FreeSWITCH-users@lists.freeswitch.org">FreeSWITCH-users@lists.freeswitch.org</a><br><a href="http://lists.freeswitch.org/mailman/listinfo/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/listinfo/freeswitch-users</a><br>UNSUBSCRIBE:<a href="http://lists.freeswitch.org/mailman/options/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/options/freeswitch-users</a><br><a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br><br><br><br><br>_______________________________________________<br>FreeSWITCH-users mailing list<br><a ymailto="mailto:FreeSWITCH-users@lists.freeswitch.org" href="mailto:FreeSWITCH-users@lists.freeswitch.org">FreeSWITCH-users@lists.freeswitch.org</a><br><a
href="http://lists.freeswitch.org/mailman/listinfo/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/listinfo/freeswitch-users</a><br>UNSUBSCRIBE:<a href="http://lists.freeswitch.org/mailman/options/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/options/freeswitch-users</a><br><a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br></div></div>
</div></body></html>