Speech Recognition Software


LumenVox provides speech recognition software. We build the speech recognition tools that you should be using in your IVR or other applications.

Speech Engine

A speech engine, or automatic speech recognizer (ASR) is the technology that allows you to interact with a computer by speaking to it. This is most often used in telephone systems, but can be applied to many other computer applications.

We've made a name for ourselves since 2001 by offering a speech engine that competes with the industry's most popular offerings. Through our competitive pricing and excellence in customer service, we're making gains on the "big guys," year after year.

LumenVox was created out of the need for a speech solution that would be accessible to everyone. The President of LumenVox, Ed Miller, found that the other providers' pricing was prohibitively expensive, so he put together a team of talented developers, speech scientists, sales and support people with the goal of creating an affordable, accurate solution.

One of our most obvious selling points is that we beat the competition's licensing costs - easily. Our nearest competitors' licenses cost more than twice what ours do.

However, we do not simply compete on price. We offer a product that is equivalent, if not superior, in accuracy. And we're always trying to squeeze point after point of accuracy out of our Speech Engine. We've done an accuracy study, which is featured on the Engine page of our website.

LumenVox Speech Engine Accuracy compared to competitors

Despite having a large competitive streak, we're still a small company. As such, we are able to take great care of our partners. It's yet another advantage that we have over the larger speech recognition providers. We pride ourselves on building lasting, fruitful partnerships. Our business development director, Gerd Graumann, is at his happiest when he's able to recommend a good match for a partner.

We've helped our customers build some great applications, too. Larry Rushing of AccelMark came to us with a program used by taxi-dispatching and other companies. We helped him rewrite some incredibly complex grammars that now allow an IVR to take taxicab requests with complete addresses.

We also helped Redmond Software improve their Spanish language OnStar application's accuracy by 42%. Offering development assistance on projects like these is just another example of how we are willing to help our partners develop impressive products.

The real star of LumenVox is our Speech Engine. We've worked hard over the years to offer a speech recognition engine which functions as accurately as the others, but remains far more affordable.

The way speech recognition works is that as the Engine takes in audio, it is guided by a list of words to be recognized called the grammar. Only the words in the grammar are considered during the conversion from audio to text, which in turn streamlines the Engine's processing. The most probable matches are what the Engine returns.

One of our most obvious selling points is that we beat the competition's 
            licensing costs - easily.

Our technology is a speaker independent solution. This means that anyone can speak to our Engine, and as long as they're saying the words the Engine is expecting to hear, they'll be understood.

It's important to distinguish this from other types of speech solutions. We do not provide dictation, nor are we a voice verification provider. Voice recognition is often used in security solutions, where the software determines who you are, not what you're saying.

We're a core technology provider, which means that our engine would be integrated into your application. Your developers can write applications using our application programming interface (API), a standards-based solution like media resource control protocol (MRCP), or using a voice platform that has already written an interface to our Engine.

We've completed interoperability testing on Avaya, Asterisk, Holly and many other speech platforms. We're always looking for more platform partners. If you're interested in testing us out, just let us know by emailing us.

Speech Tuner

Finally, our Windows-based Speech Tuner is another tool that we're quite proud of. It gives you the ability to fine-tune and maintain your speech applications. You can transcribe call data, make on-the-fly Engine changes and then test those new settings instantly for effectiveness with recorded calls. We have extensive documentation on the Speech Tuner on our Support site.

We hope that you take a moment to reacquaint yourselves with our products and services if you're familiar with us, or start a new relationship if you don't already know us. Either way, we hope to work with you soon.

© 2018 LumenVox, LLC. All rights reserved.