Abstract: An effective and simple psychoacoustic bass generation system generates a harmonic signal having inter-modulation controllable to remain below a threshold level and includes a high-pass filter configured to pass harmonics which are reproducible with fidelity by the loudspeaker or other transducer and a loudness matching block configured to compensate the loudness of the desired harmonics to match the loudness of the original signal.

Abstract: An effective and simple psychoacoustic bass generation system generates a harmonic signal having inter-modulation controllable to remain below a threshold level and includes a high-pass filter configured to pass harmonics which are reproducible with fidelity by the loudspeaker or other transducer and a loudness matching block configured to compensate the loudness of the desired harmonics to match the loudness of the original signal.

Abstract: In embodiments of this invention howling sounds are removed without disturbing speech and music. Microphone signals are converted to the frequency domain where howling frequencies are detected and estimated using howling to noise ratios. A notch filter adaptively designed for the howling frequency is applied to remove the howling.

Abstract: A universal reconfigurable system and method are provided for reducing nonlinear echo and residual echo, and for echo leakage prevention in various time-varying and complex environments is proposed in this invention. According to one embodiment an echo cancellation system includes (1) an adaptive linear filter implemented in either time-domain or frequency-domain; (2) a nonlinear echo suppression; (3) echo leakage prevention; (4) direct current (DC), low frequency residual echo and noise reduction; (5) time-domain nonlinear processor (NLP) to reduce the residual echo; and (6) frequency-domain NLP to further reduce the residual echo. The echo cancellation system is universally applicable to acoustic echo cancellation (AEC) and/or electrical echo cancellation applications. In terms of AEC, this invention is reconfigurable for one or more microphones and/or one or more reference channels. The numbers of microphones and reference channels are user configurable.

Abstract: A system and method are disclosed for effective and simple stereo widening over arbitrarily-configured speakers and its real-time implementation. According to one embodiment, the system includes five processing units: (1) elevation processing unit; (2) side signal or difference signal processing unit; (3) center signal processing unit; (4) binaural signal processing unit; and (5) stereo limiter (to prevent the clipping) unit. Any of the five processing units may be included or omitted depending on the application. In general, embodiments of the present invention provide effective and simple stereo widening schemes with good audio quality results, but without having substantial requirements on the speakers' positional configuration. For example, the proposed schemes can apply to both symmetric and non-symmetric stereo loudspeakers.

Abstract: In embodiments of this invention howling sounds are removed without disturbing speech and music. Microphone signals are converted to the frequency domain where howling frequencies are detected and estimated using howling to noise ratios. A notch filter adaptively designed for the howling frequency is applied to remove the howling.

Abstract: An effective and simple psychoacoustic bass generation system generates a harmonic signal having inter-modulation controllable to remain below a threshold level and includes a high-pass filter configured to pass harmonics which are reproducible with fidelity by the loudspeaker or other transducer and a loudness matching block configured to compensate the loudness of the desired harmonics to match the loudness of the original signal.

Abstract: A system and method are disclosed for effective and simple stereo widening over arbitrarily-configured speakers and its real-time implementation. According to one embodiment, the system includes five processing units: (1) elevation processing unit; (2) side signal or difference signal processing unit; (3) center signal processing unit; (4) binaural signal processing unit; and (5) stereo limiter (to prevent the clipping) unit. Any of the five processing units may be included or omitted depending on the application. In general, embodiments of the present invention provide effective and simple stereo widening schemes with good audio quality results, but without having substantial requirements on the speakers' positional configuration. For example, the proposed schemes can apply to both symmetric and non-symmetric stereo loudspeakers.

Abstract: A universal reconfigurable system and method are provided for reducing nonlinear echo and residual echo, and for echo leakage prevention in various time-varying and complex environments is proposed in this invention. According to one embodiment an echo cancellation system includes (1) an adaptive linear filter implemented in either time-domain or frequency-domain; (2) a nonlinear echo suppression; (3) echo leakage prevention; (4) direct current (DC), low frequency residual echo and noise reduction; (5) time-domain nonlinear processor (NLP) to reduce the residual echo; and (6) frequency-domain NLP to further reduce the residual echo. The echo cancellation system is universally applicable to acoustic echo cancellation (AEC) and/or electrical echo cancellation applications. In terms of AEC, this invention is reconfigurable for one or more microphones and/or one or more reference channels. The numbers of microphones and reference channels are user configurable.