Submenu

 Go back to the
 previous page


Related links

 ADM - Adaptive Dual
 Microphone

 AEC - Acoustic
 Echo Cancelling

 AFR - Acoustic
 Feedback Reduction

 AVQ - Automatic
 Volume & eQualization
 Control

 AVQ-Me - AVQ for Music
 and Entertainment

 AWNR - Adaptive Wind
Noise Reduction

 DRC - Dynamic Range
 Compression

 EasyListen - speech
 slowing-down

 Mono to Stereo
 conversion

 MuRefiner

 NS - Noise Suppression

 PLC - Packet
 Loss Concealment

 ISM - Intelligent
 Speech Mixer

 Vocoders

 White papers

 Terminology index

 

EasyListen™

You are here: Technologies -> EasyListen

About EasyListen™

Real time fast speech intelligibility enhancer

As time passes, it seems that people tend to talk faster and faster each year. Some of them even tend to speak fast to us in foreign language. We often don’t have enough time to catch phone numbers, dates or other important information. And they don’t have time to repeat …

Finally there is a technology that can help. EasyListen™ dynamically slows down the incoming speech making it natural and understandable. While “time stretching” technologies for speech modification are known for a long time (see here for example), EasyListen is more complicated and intelligent than classical approaches. The key is the word “dynamically” and here are the main differences:

  • EasyListen is a real time technology. Slowing down an audio signal in real time means a delay is accumulating fast. In any real voice conversation such accumulated delay must be compensated at some moment without missing important parts of the speech. Pauses happening in each real conversation and shortening such pauses is used to reduce the delay. But what if a person continuously speaks fast with no interrupts – in this case EasyListen will stop slowing his/her monolog when the maximal allowed delay is exceeded.
  • EasyListen does not stretch tones, noises and other irrelevant sounds
  • EasyListen does not slow down all parts of the speech equally. It performs a sophisticated speech analysis and stretches only those parts that are important for understanding but too fast and only to the extent it is necessary. Consequently, different vowels, plosives and fricatives are not stretched in an equal manner. And some sounds that that do not require stretching are not processed at all. For example such continuous sounds as “Aaaa”, “yee”, “Mmm” etc are not stretched.

EasyListen technology is language independent and works equally well with Western and Eastern tonal languages.

Applications of EasyListen technology include:

  • Consumer voice communication (mobile and cordless phones, Bluetooth headsets)
  • Professional voice communication (e.g. communication between foreign pilots and control tower)
  • TV and radio receivers
  • Voice recorders
  • Voice mailboxes

Hearing is believing - interactive audio examples

Use control buttons on the interactive screenshots to play/stop/navigate.

Example 1.
Original speech recording - a phone dialog between two fast talkers:


EasyListen™ output - while the length of the dialog has not increased significantly, the intelligibility has certainly improved:



Example 2.
Original speech recording - a presentation example; the presenting lady probably speaks too fast for most people:


EasyListen™ output - slowing her down by 15% (on average), makes the content much more intelligible:


 

Please, contact us for more information.

 

 

go back to the Technologies page.

 
 
 
   

(c) Alango Technologies Ltd., 2006-14

Site development: http://radzishevsky.info, 2006-8