Hughes Electronics Corporation

Hughes Aircraft Company

Hughes Electronics Corporaton

Prabhat K Gupta

Sanjay Gupta

Allan B Lamkin

Kalyan Ganesan

Kumar Swaminathan

Adrian J Morris

Shrirang Jangi

W Robert Kepley

Bob Kepley

Walter Robert Kepley, Iii

Paul Gaske

Ho Lee

W Robert Kepley, Iii

P Vijay Kumar

The invention describes a method and device of comfort noise generation for echo cancelers. The invention stores the actual background noise of a channel in a noise buffer. When the echo suppressor is activated by the echo canceler to suppress any unwanted signals on the channel, the buffer provides the stored noise to the channel to provide continuous background noise. To ensure that neither speech nor impulse noise enters the buffer, hangover delays and energy threshold levels are provided such that signals within the delay period or signals which exceed the energy threshold level are screened from entering the buffer.

Comfort noise generator for echo cancelers

A method of detecting double talk in an echo canceller is provided. The method includes determining far end, near end and echo signal cancelled power, determining the presence of near end, maybe near end and far end speech, and controlling an echo suppressor and filter coefficient updates in an echo canceller.

Method of detecting double talk in an echo canceller

A signal classifier and a echo suppressor for a facsimile calls in an echo prone network. The signal classifier quickly and reliably differentiates facsimile calls from voice and modem calls. The echo suppressor suppresses echo and has a variable hangover time which avoids attenuating and distorting a return transmission signal.

Echo suppressor for facsimile calls in a satellite network

A bit rate Codebook Excited Linear Predictor (CELP) communication system which includes a transmitter that organizes a signal containing speech into frames of 40 millisecond duration, and classifies each frame as one of three modes: voiced and stationary, unvoiced or transient, and background noise.

Voiced, unvoiced or noise modes in a CELP vocoder

A method of detecting narrow-band signals in an echo canceller to control the coefficient update process, echo suppression process, and the disablement of the echo canceller is provided. The method includes the steps of searching for the presence of a generic tone, searching for DTMF tones, searching for a 2100 Hz tone and, if a generic tone is found, searching for a dial tone. If a generic tone is detected on the far end signal, a tone flag is set which triggers the disablement of the echo canceller coefficient update process and the enablement of the echo suppressor process. If a 2100 Hz tone is detected, the echo canceller is disabled. If a DTMF signal is detected, DTMF flags are set which are used by the echo suppressor for suppressing the near end echo. Finally, if a dial tone is detected, dial tone flags are set which cause the disablement of the echo suppressor.

Method of detecting narrow-band signals in an echo canceller

An improved error control coding scheme is implemented in low bit rate coders in order to improve their performance in the presence of transmission errors typical of the digital cellular channel. The error control coding scheme exploits the nonlinear block codes (NBCs) for purposes of tailoring those codes to a fading channel in order to provide superior error protection to the compressed half rate speech data. For a half rate speech codec assumed to have a frame size of 40 ms, the speech encoder puts out a fixed number of bits per 40 ms. These bits are divided into three distinct classes, referred to as Class 1, Class 2 and Class 3 bits. A subset of the Class 1 bits are further protected by a CRC for error detection purposes. The Class 1 bits and the CRC bits are encoded by a rate 1/2 Nordstrom Robinson code with codeword length of 16. The Class 2 bits are encoded by a punctured version of the Nordstrom Robinson code. It has an effective rate of 8/14 with a codeword length 14. The Class 3 bits are left unprotected. The coded Class 1 plus CRC bits, coded Class 2 bits, and the Class 3 bits are mixed in an interleaving array of size 16.times.17 and interleaved over two slots in a manner that optimally divides each codeword between the two slots. At the receiver the coded Class 1 plus CRC bits, coded Class 2 bits, and Class 3 bits are extracted after de-interleaving.

High performance error control coding in channel encoders and decoders

A method of synthesizing an echo path in an echo canceller is provided. The method includes the steps of obtaining near end and far end signal samples at an echo canceller, locating the echo in a background task using a whitened version of the signal, transferring the echo location information to a foreground task, and eliminating the echo in the incoming signals by filtering the signal in the location passed to the foreground task by the background task.

Method for synthesizing an echo path in an echo canceller

A voice activity detector (VAD) which determines whether received voice signal samples contain speech by deriving parameters measuring short term time domain characteristics of the input signal, including the average signal level and the absolute value of any change in average signal level, and comparing the derived parameter values with corresponding thresholds, which are periodically monitored and updated to reflect changes in the level of background noise, thereby minimizing clipping and false alarms.

Voice activity detector for speech signals in variable background noise

The perception of speech processed by a CELP based coder, such as a VSELP coder, when operating in noisy background conditions is improved by removing swirl artifacts during silence periods. This is done by removing the low frequency components of the input signal when no speech is detected. A speech activity detector distinguishes between a periodic signal, like speech, and a non-periodic signal, like noise by using most of the VSELP coder internal parameters to determine the speech or non-speech conditions. To prevent the VSELP coder from determining pitches for non-periodic signals, a high pass filter is applied to the input signal to remove the pitch information for which the VSELP coder searches.

Removal of swirl artifacts from celp-based speech coders

A method for encoding a signal that includes a speech component is described. First and second linear prediction windows of a frame are analyzed to generate sets of filter coefficients. First and second pitch analysis windows of the frame are analyzed to generate pitch estimates. The frame is classified in one of at least two modes, e.g. voiced, unvoiced and noise modes, based, for example, on pitch stationarity, short-term level gradient or zero crossing rate. Then the frame is encoded using the filter coefficients and pitch estimates in a particular manner depending upon the mode determination for the frame, preferably employing CELP based encoding algorithms.

Mode-specific method and apparatus for encoding signals containing speech

Noise is suppressed during pauses and silent periods in conversation, and voiced signals are freely passed in a mobile communications system thereby improving the quality of the transmitted speech. Noise suppression is implemented in such a way that a low level of noise is still allowed to pass to provide presence of the remote speaker; that is, the line is not made completely silent as this may falsely indicate that the connection has been interrupted. The noise suppression has an added feature of decreasing the background noise fractionally when voice is no longer detected. This provides perceptually improved quality of the communication.

Noise suppression in digital voice communications systems

A codec uses low cost digital signal processors (DSPs) to implement the codebook excited linear prediction (CELP) algorithm. The flexible architecture provides a platform for implementing a family of CELP codecs. In a specific example, an 8 Kbps CELP codec is partitioned into parallel tasks for real time implementation on dual DSPs with flexible intertask communication, prioritization and synchronization with asynchronous transmit and receive frame timings. The two DSPs are used in a master-slave pair. Each DSP has its own local memory. The DSPs communicate to each other through interrupts. Messages are passed through a dual port RAM. Each dual port RAM has separate sections for command-response and for data.

Real-time implementation of a 8Kbps CELP coder on a DSP pair

A voice activity detector (VAD) which determines whether an input signal contains speech by deriving parameters measuring short term time domain characteristics of the input signal, including the average signal level and the absolute value of any change in average signal level, and comparing the derived parameter values with corresponding predetermined threshold values. In order to further minimize clipping and false alarms, the VAD periodically monitors and updates the threshold values to reflect changes in the level of background noise.

Growing community of inventors

Germantown, MD, United States of America

Prabhat K Gupta