Abstract: A packet voice communication system having a jitter buffer external to a voice processor. The jitter buffer stores voice packets received from a packet network. The voice processor processes the voice packets from the jitter buffer. A jitter buffer processor may place an indicator in each voice packet it holds. The indicator can indicate a length of time the voice packet was held. The rate at which packets come from the jitter buffer may be based upon the indicator, a higher rate if holding times are high and a slower rate if low. The voice processor can store the voice packets in a packet queue prior to processing the voice packets. The rate voice packets come to the voice processor may be based upon how full the packet queue is, a higher rate if the packet queue is relatively empty and a slower rate if relatively full.

Abstract: A technique for suppressing non-stationary noise, such as wind noise, in an audio signal is described. In accordance with the technique, a series of frames of the audio signal is analyzed to detect whether the audio signal comprises non-stationary noise. If it is detected that the audio signal comprises non-stationary noise, a number of steps are performed. In accordance with these steps, a determination is made as to whether a frame of the audio signal comprises non-stationary noise or speech and non-stationary noise. If it is determined that the frame comprises non-stationary noise, a first filter is applied to the frame and if it is determined that the frame comprises speech and non-stationary noise, a second filter is applied to the frame.

Abstract: System and method for processing communication signals in a communication system having a detector for detecting a parameter of a communication signal. A communication signal is provided to a non-linear processor (NLP) adapted to examine the signal and to decide whether or not to enter an active state based upon a parameter of the signal. If the NLP enters an active state, the NLP performs non-linear processing on the signal. The NLP communicates to the detector whether the NLP is active or inactive. If the NLP is active, a processing step of the detector is disabled.

Abstract: An apparatus and methods supporting improved echo cancellation in a packet network is disclosed. An embodiment of the present invention may permit a split-band communication system operating in a narrowband mode to provide improved suppression of hybrid echo generated by the conversion of signals from a packet network to a two-wire analog circuit. By detecting the presence of signal characteristics outside of the relatively narrower bandwidth to be communicated, such an embodiment may more accurately detect the occurrence of speech received from a party on a relatively wider bandwidth communication circuit. The accurate detection of speech permits more effective suppression of any hybrid echo remaining after echo cancellation. An embodiment of the present invention may also have application in other systems that detect signals received via a path having a bandwidth greater than that to be transmitted.

Abstract: A beamforming technique for a microphone array is described to attenuate a source of undesired noise that is deemed the most limiting to audio quality in an acoustic environment. Possible sources of undesired noise include echo, background noise (stationary) and other interference signals (non-stationary). The beamforming technique is updated based on long-term evaluations. Once an evaluation occurs and a decision is made, the beamformer adapts with a maximum responsiveness and without intentional delay, and therefore not affecting the beamformer's tracking ability. When fixed beamforming is utilized, one of several fixed beamformers having different attenuation targets are selected to implement noise attenuation. When adaptive beamforming is utilized, the beamformer adapts whenever the selected target is deemed dominant.

Abstract: Methods, systems, and computer program products for enhancing transmitted audio signals for a video conference call are disclosed. Facial recognition technology and audio processing techniques determine a position of a speaking member of a party on a video conference. Based on the determined position of the speaking member, the directionality of the received audio signals can be adjusted towards the speaking member. The video conference system can transmit the adjusted audio signals.

Abstract: During playing of a streamed audio while engaging in a voice communication session, a communication device may be operable to generate a background audio during an echo cancellation operation for the voice communication session, using known information associated with the streamed audio. The streamed audio may comprise audio content that is received from one or more external sources and/or locally generated within the communication device. The known information associated with the streamed audio may comprise an echo replica corresponding to an echo associated with the streamed audio and/or a modified version of the streamed audio. In instances when a residual echo suppression is operational during the echo cancellation operation, the communication device may be operable to suppress a residual echo based on a signal level of the echo replica corresponding to the echo associated with the streamed audio and/or a signal level of the modified version of the streamed audio.

Abstract: Certain aspects of a method and system for a dual mode subband acoustic echo canceller with integrated noise suppression may include splitting an input signal into a lowband component and a highband component. The subbands of each of the lowband component and the highband component may be processed in order to reduce an echo associated with the input signal and to suppress the noise associated with the input signal.

Abstract: Method of processing a transmitted encoded media data stream is received. If a data element arrives prior to, or at, a predetermined playout deadline, the data element is decoded, the media represented by the decoded data element is played, and the data element is provided to a decoder state machine to update a decoder state. If a data element arrives after the predetermined playout deadline, the data element is provided to the decoder state machine to update the decoder state. In one embodiment, if the specified data element fails to arrive by the playout deadline, a subsequently received data element is saved in memory. Then, if the specified data element arrives after the predetermined playout deadline, the specified data element and the saved, subsequently received, data element are provided to the decoder state machine to update the decoder state.

Abstract: A method and system for improving speech quality may include estimating at least one component of a distorted portion of a speech signal from at least one component of an undistorted portion of the speech signal and reinforcing the component of the distorted portion based on the estimating. The components may include the pitch, spectral envelope and spectral energy of the speech signal. The undistorted portion of the speech signal may be delayed and the components of the distorted portion may be interpolated from the components of a delayed undistorted portion and a current undistorted portion of the speech signal. The components of the distorted portion of the speech signal may be extrapolated from a current undistorted portion of the speech signal. Components of the distorted portion of the speech signal may be estimated from frequency bands other than the frequency band affected by the distortion.

Abstract: Certain aspects of a method and system for a dual mode subband acoustic echo canceller with integrated noise suppression may include splitting an input signal into a lowband component and a highband component. The subbands of each of the lowband component and the highband component may be processed in order to reduce an echo associated with the input signal and to suppress the noise associated with the input signal.

Abstract: A system and method for provide a stable gain from an adaptive gain control device in a signal path. An echo canceller is also located in the signal path, and is used to provide performance information regarding losses in the signal. This performance information is fed to the automatic gain control device via a connection. The automatic gain control device thereafter uses the performance information to determine a maximum gain that might be provided based upon losses cause by echo conditions. The gain however is limited in order to provide for a stable system. The performance information includes a loss rate that includes a combination of the echo return loss and the echo return loss enhancement.

Abstract: Methods for using communication network statistics in the operation of a real-time communication system are disclosed. Embodiments of the invention may provide improved playback of real-time media streams by incorporating into the algorithms used for playback of the media stream network statistics typically calculated by some transport protocols. An additional aspect of the present invention may include machine-readable storage having stored thereon a computer program having a plurality of code sections executable by a machine for causing the machine to perform the foregoing.

Abstract: System and method for processing communication signals in a communication system having a detector for detecting a parameter of a communication signal. A communication signal is provided to a non-linear processor (NLP) adapted to examine the signal and to decide whether or not to enter an active state based upon a parameter of the signal. If the NLP enters an active state, the NLP performs non-linear processing on the signal. The NLP communicates to the detector whether the NLP is active or inactive. If the NLP is active, a processing step of the detector is disabled.

Abstract: Certain aspects of a method and system for a dual mode subband acoustic echo canceller with integrated noise suppression may include splitting an input signal into a lowband component and a highband component. The subbands of each of the lowband component and the highband component may be processed in order to reduce an echo associated with the input signal and to suppress the noise associated with the input signal.

Abstract: A system and method for provide a stable gain from an adaptive gain control device in a signal path. An echo canceller is also located in the signal path, and is used to provide performance information regarding losses in the signal. This performance information is fed to the automatic gain control device via a connection. The automatic gain control device thereafter uses the performance information to determine a maximum gain that might be provided based upon losses cause by echo conditions. The gain however is limited in order to provide for a stable system. The performance information includes a loss rate that includes a combination of the echo return loss and the echo return loss enhancement.

Abstract: Systems and methods are described that automatically adjust a value of a parameter relating to the delivery of media content, such as audio content or image content, based on both environmental conditions and on automatically-learned user preference data. For example, a first embodiment adjusts a volume setting used to control the delivery of an audio signal based both on environmental noise conditions and upon automatically-learned user preference information, wherein the user preference information is derived by monitoring user-implemented adjustments to the volume setting after application of an automatic adjustment thereto.

Abstract: Techniques for loudspeaker localization are provided. Sound is received from a loudspeaker at a plurality of microphone locations. A plurality of audio signals is generated based on the sound received at the plurality of microphone locations. Location information is generated that indicates a loudspeaker location for the loudspeaker based on the plurality of audio signals. Whether the generated location information matches a predetermined desired loudspeaker location for the loudspeaker is determined. A corrective action with regard to the loudspeaker is enabled to be performed if the generated location information is determined to not match the predetermined desired loudspeaker location for the loudspeaker.

Abstract: System and method for processing communication signals in a communication system having a detector for detecting a parameter of a communication signal. A communication signal is provided to a non-linear processor (NLP) adapted to examine the signal and to decide whether or not to enter an active state based upon a parameter of the signal. If the NLP enters an active state, the NLP performs non-linear processing on the signal. The NLP communicates to the detector whether the NLP is active or inactive. If the NLP is active, a processing step of the detector is disabled.

Abstract: Methods and apparatus of managing a communication system, wherein a decision regarding a level of activity at a first end is made based at least in part on the level of activity at the second end. In one embodiment, the energy level of a first-end audio signal is measured. The first end is declared voice-active if the first-end energy level is greater than or equal to a first threshold value. The first end is declared voice-inactive if the first-end energy level is less than the first threshold value. To determine the value of the first threshold value, the energy level of a second-end audio signal is measured. If the second-end energy level is greater than or equal to a second threshold value, the second end is declared voice-active, in which case the first threshold is maintained at a relatively high level. If the second-end energy level is less than the second threshold value, the second end is declared voice-inactive, in which case the first threshold is maintained at a relatively lower level.