Video
Speech Recognition Enhancement

Video  / Speech Recognition Enhancement
Aug.19, 2018

Alango Voice Activity Detection (VAD) is an embedded DSP-based technology aimed to discriminate between the human speech and non-speech signal. The stand-alone VAD consumes very little power and allows “heavier” signal processing tasks (such as speech recognition) to cease activity and the DSP to sustain a low-power consumption mode as long as human voice activity is not detected. Alango VAD technology implements a proprietary high resolution spectrum noise estimation algorithm allowing reliable voice detection and keeping false detection rate low.

Feb.12, 2018

This is a demonstration of Alango’s far-field speech enhancement algorithms and software. Typical applications include smart-home devices and smartphones. The demo features a voice-controlled stereo music player with a 6-microphone linear array simulating a voice controlled TV. The software integrates Alango's proprietary Voice Enhancement Package of software technologies including stereo echo cancellation and far-field beamforming. The output signal from the Voice Enhancement Package is fed into Sensory speech recognition software used to interpret user commands.

Jan.7, 2018

Alango showcases its 6-microphone linear array beamforming. A voice-controlled music player recognizes the trigger word while Alango algorithm enhances the recognition.

Nov.19, 2017

Real World Application of Alango’s Voice Enhancement Package in Home Robot.

Oct.23, 2017

This is a comparison of Voice Trigger recognition sensitivity and robustness of Amazon Echo and Alango’s Voice Enhancement Package (VEP) demo combined with Sensory Voice Trigger technology. Typical VEP applications include smart-home devices and smartphones. The output, enhanced, noise and echo free signal from the Voice Enhancement Package is fed into Sensory Voice Trigger technology.

Aug.2, 2017

This presentation was given by Dr.Alexander Goldin, CEO of Alango Technologies, at Afeka Conference for Speech Processing on July 3rd, 2017. Alexander gave an overview of technologies for speech recognition enhancement in Smart Home devices. Alango has been working in this field for several years, but the accent of the presentation was given not on Alango technologies, but on the integration of speech enhancement and speech recognition technologies from different vendors for the best system performance. The original video as well as other interesting presentations can be found here http://www.afekaconference.co.il/sp2017/

Mar.12, 2017

This is a demonstration of Alango’s far-field speech enhancement algorithms and software. Typical applications include smart home devices and smartphones. The demo features a voice-controlled stereo music player with a built-in four-microphone array. The software integrates Alango's proprietary Voice Enhancement Package of software technologies including stereo echo cancellation and far-field beamforming. The output signal from the Voice Enhancement Package is fed into Sensory speech recognition software used to interpret user commands.

Jan.3, 2017

This is a demonstration of Alango’s far-field speech enhancement algorithms and software. Typical applications include smart home devices and smartphones. The demo features a voice-controlled stereo music player with a built-in four-microphone array. The software integrates Alango's proprietary Voice Enhancement Package of software technologies including stereo echo cancellation and far-field beamforming. The output signal from the Voice Enhancement Package is fed into Sensory speech recognition software used to interpret user commands.

Aug.8, 2016

Robert Schrager, sales and marketing director at Alango Technologies Ltd. demonstrates application of Alango proprietary, dual microphone digital noise reduction technology for smart watch application. Dual microphone noise reduction in smart watches poses a special challenge due to uncertainty of user’s voice direction (left, right, front, back) with omnipresent ambient noise. The video clip shows that noises can be successfully removed with user’s voice attenuation and distortion.

Dec.10, 2014

This clip will demonstrate the powerful combination of the advanced stereo echo canceller by Alango, robust speech recognition by Sensory and powerful digital signal processing available with CEVA DSP. Now you can control your home media centre by voice, hands-free and hustle-free.

Top ▲