Echo cancellation for a two-way audio communication includes receiving, at an AEC system from microphone(s), an audio signal based on, at least in part, near-end signals and reproduced far-end signals. Loudspeaker(s) reproduced the far-end signals. The AEC system is operated, at least in part, with filter(s) so as to update estimates of coefficients of an acoustic channel from the loudspeaker(s) to the microphone(s). Control parameter(s) affecting an operation of the AEC system that is/are configurable and is/are set to value(s), from a range of values, is/are determined, based on estimating an accuracy of the estimates of the coefficients of the acoustic channel and a characteristic of the near-end signals. The AEC system controls the filter(s) with different values of the control parameter(s) at different times.

The present disclosure relates to a device having an output transducer and an input transducer, wherein the device is configured to be operated so as to reduce or cancel signal content that is output via the output transducer in the input transducer signal.

The present invention discloses a method of noise reduction for an intelligent network communication, which includes the following steps: first, receiving a local sound message through a sound receiver of a communication device at the transmitting end. Next, a voice recognizer is used to identify the voice characteristics of the speaker; then, it is determined from a voice database whether there is a corresponding or similar voice characteristic of the speaker recognized by the voice recognizer. Finally, filtering other signals other than the voice characteristic signal of the speaker through a sound filter to obtain the original sound emitted by the speaker.

A method and an electronic device for detecting crosstalk data are provided. The method for detecting crosstalk data can detect whether an audio data stream includes crosstalk data. The method includes: receiving a first audio data block, a second audio data block, and a reference time difference, wherein the first audio data block and the second audio data block separately include a plurality of audio data segments; using a time difference between an acquisition time of an audio data segment in the first audio data block and a corresponding audio data segment in the second audio data block as an audio segment time difference; and determining that the audio data segment of the first audio data block includes crosstalk data when the audio segment time difference does not match the reference time difference.

A voice communication apparatus includes a communication unit configured to communicate with one or more another terminals via a network, a voice signal processing unit configured to acquire a first voice signal collected from a voice input terminal, acquire a second voice signal output from another terminal, and detect whether there is howling based on the first and second voice signals, a control unit configured to determine whether a device connected to the voice input terminal or a device connected to the voice output terminal is a howling cause based on a detection result of the voice signal processing unit, and an alert notification unit configured to generate and output an alert screen when the control unit determines that the device connected to the voice input terminal or the device connected to the voice output terminal is the howling cause.

A multi-channel acoustic echo cancellation (AEC) system that includes a step-size controller that dynamically determines a step-size value for each channel and each tone index on a frame-by-frame basis. The system determines that near-end signals are present by calculating a scaled error and determining that the scaled error exceeds a threshold value. When the scaled error exceeds the threshold value, the system may switch from a first cost function to a second cost function and determine a step-size value using a robust algorithm. The robust algorithm may prevent the system from diverging due to the presence of the near-end signal. For example, the robust algorithm may select a different cost function to determine the step-size value and/or combine different step-size computations, resulting in the step-size value being temporarily reduced. Thus, the robust algorithm may enable the AEC to better model the near-end disturbance statistics while the near-end signal is present.

A conference system includes a conversation state determiner that determines whether or not the state of first and second users is a direct conversation state in which direct conversation is possible without using a speech system, and an output controller that controls whether or not to cause the speech system to output a first acquired voice from a second speaker, based on the determination result of the conversation state determiner.

Systems and methods are described to enable a device of a user to automatically join an ongoing conference, where the device is not currently joined to the conference. A first audio signature is generated based on voices of users already in the conference, and a second audio signature is generated based on an audio signal captured by a microphone of the device associated with the first user when the device associated with the first user was not joined to the conference. The first audio signature and the second audio signature are compared, and in response to determining that first audio signature matches the second audio signature, the device associated with the first user is joined to the conference.

Method and System for avoiding howling disturbance especially on conferences, wherein the method comprising the steps of using a howling detector unit implemented inside a multipoint control unit to receive an audio stream input from a client, analyzing the audio input with the howling detector in order to verify if howling noise is present, using at least two of a skewness analysis, a flatness analysis, a crest analysis, a rolloff analysis and preventing the audio stream input to be forwarded as an output to an audio mixer, if howling noise is present.

A conference terminal and an echo cancellation method for a conference are provided. In the echo cancellation method, a synthetic speech signal is received. The synthetic speech signal includes a user speech signal of a speaking party corresponding to a first conference terminal of multiple conference terminals and an audio watermark signal corresponding to the first conference terminal. One or more delay times corresponding to the audio watermark signal are detected in a received audio signal. The received audio signal is recorded through a sound receiver of a second conference terminal of the conference terminals. An echo in the received audio signal is canceled according to the delay time.