Pre-selectable and dynamic configurable multistage echo control system for large range level of acoustic echo

Abstract

A method for controlling echo of a voice recognition system includes receiving at least two audio input signals corresponding to sound sensed by at least two microphones in a physical space. A first audio input signal of the at least two audio input signals is received on a primary channel, and each remaining audio input signal is received through a respective secondary channel. The method includes selecting, by a processor, based on an echo power level of a speaker in the physical space, a subset of echo control functions (ECFs) from among a plurality of ECFs of a multistage echo control system. Each ECF modifies the at least two audio input signals to reduce echo. The method includes generating a corresponding number of audio output signals by processing the signals received on the primary and secondary channels through the selected subset of ECFs, and outputting the audio output signals.

Description

BACKGROUND1. Technical Field[0001]The present disclosure generally relates to electronic devices with audio speakers and microphones, and more particularly to acoustic echo cancellation systems for voice signal processing by such electronic devices.2. Description of the Related Art[0002]Mobile devices such as, smart phones, tablets, laptops or other electronic devices, oftentimes include voice recognition (VR) systems to provide hands free voice control of the device. Although VR technologies have been improving, accurate voice recognition remains a technical challenge. For example, during device playback, users try to interact with the device by saying a trigger phase or command to instruct the device to pause the playback and perform another task. The user expects the VR engine to recognize the user's speech, distinguish the user's speech from the playback, and to respond quickly and correctly. However, in part due to the small form factor of the mobile device, the echo signal det...

AI Technical Summary

Benefits of technology

The patent text describes a method, electronic device, and computer program product for controlling echo in a voice recognition system. The technical effect of this patent is to improve the accuracy of voice recognition in electronic devices with audio speakers and microphones, by canceling out the echo caused by the speaker and reducing the impact of the echo signal on the user's speech signal. This is achieved by selecting a subset of echo control functions based on the echo power level of the speaker and generating a corresponding number of audio output signals through the selected set of echo control functions.

Problems solved by technology

Although VR technologies have been improving, accurate voice recognition remains a technical challenge.

Consequently, there is a problem in that the user's speech signal is masked by the echo signal (i.e., the speech to echo ratio (SER) can be more than −30 dB), making it difficult to be heard by the mobile device.

There is a problem in that a conventional acoustic echo canceller (AEC) of a mobile device cannot cancel an SER that is greater than 30 dB, and thus, the speech of a user is often unrecognized or mis-recognized by the VR engine.

Images