San Diego, CA (JUNE 1, 2010) — Users can now deliver outbound messages reliably by detecting answering machine and voicemail tones at greater than 90 percent accuracy with LumenVox's answering machine detection.
Available in the LumenVox Speech Engine release 9.5, the new answering machine detection is an effective method to tell if an outbound call has been connected to a live person or an automated recording system.
The technology is compatible with most voice platforms and PBX systems, and is fully supported through the standards–based Media Resource Control Protocol (MRCP). Since MRCP is supported by almost every major voice platform, it makes plugging answering machine detection into your application simple.
"This technology is a welcome addition to the suite of speech recognition software offered by LumenVox," said LumenVox President Edward Miller. "When combined with the Speech Engine, it offers a complete solution for outbound messaging applications."
To enable answering machine detection, customers simply load a special grammar file that tells the Speech Engine to listen for beeps in the audio. As soon as it detects a tone, the Engine returns the event to the application or platform.
LumenVox's tone–based detection works differently than many other answering machine detection technologies. It listens for the actual tones, which allows it to achieve very high accuracy compared to the normal energy–based detection.
Since LumenVox is listening for the beep, an outbound dialing application will not know whether it is connected to a voicemail system for the beginning of the call.
The guiding principle behind the technology is that an outbound application can treat both humans and answering machines the same up until the moment the beep plays. To implement this technology, customers perform the following steps:
"In addition to being simple, this approach is very reliable," Miller said. "Our testing indicates that we achieve around 95% accuracy in detecting tones."
Other answering machine detection systems attempt to measure the duration of the called party's speech to determine whether it is a live person or a machine, but this is an imprecise method prone to failure. LumenVox's method is simple and effective.
LumenVox's answering machine detection technology runs on any system supported by the LumenVox Speech Engine, version 9.5 or newer. This means it can run on any modern Windows release or supported Linux distribution.
Answering machine detection is supported over MRCP, both versions 1 and 2. It should be compliant with any VoiceXML platform that also supports MRCP.
Specific answering machine detection licenses are needed in order to perform the tone detection.
System resource requirements are relatively minimal, but may scale up for very large numbers of simultaneous detections. This extra system load is in addition to any speech recognitions the Speech Engine may be doing.
© 2016 LumenVox, LLC. All rights reserved.