Abstract: A solution is proposed for monitoring quality of a conference call. Corresponding methods are implemented on a client computing machine of a participant to the conference call, a server computing machine for managing the conference call and a computing system comprising the server computing machine and a plurality of client computing machines, respectively. One or more test signals (having corresponding test frequencies within a human audible range) are generated by each client computing machine in the mute mode without audible effects on the participant. The test signals are transmitted from the client computing machine to the server computing machine. The server computing machine verifies the quality of the conference call for the participant according to a comparison of one or more received signals (received from the client computing machine) with the test signals. Computer programs and computer program products for performing the corresponding methods are also proposed.

Abstract: Systems and methods for call detail record (CDR) analysis to determine a risk score for a call and identify fraudulent activity and for fraud detection in Interactive Voice Response (IVR) systems. An example method may store information extracted from received calls. Queries of the stored information may be performed to select data using keys, wherein each key relates to one of the received calls, and wherein the queries are parallelized. The selected data may be transformed into feature vectors, wherein each feature vector relates to one of the received calls and includes a velocity feature and at least one of a behavior feature or a reputation feature. A risk score for the call may be generated during the call based on the feature vectors.

Abstract: The present invention is relative to a method for analyzing an signal (INS), representative of a wave that propagates in a physical medium, providing characteristic parameters of said signal, said method being implemented on a computing platform (CP), requiring only fixed point computations, and with a reduced number of multiplications. Parameters that are provided can be one or several of the following: instantaneous phase (IP), instantaneous amplitude (IA), instantaneous frequency (IF), Sliding Fourier Transform (STFT).

Abstract: The disclosure provides a method and an apparatus for detecting a voice activity in an input audio signal composed of frames. A noise characteristic of the input signal is determined based on a received frame of the input audio signal. A voice activity detection (VAD) parameter is derived based on the noise characteristic of the input audio signal using an adaptive function. The derived VAD parameter is compared with a threshold value to provide a voice activity detection decision. The input audio signal is processed according to the voice activity detection decision.

Abstract: A system and method are presented for acoustic echo cancellation. The echo canceller performs reduction of acoustic and hybrid echoes which may arise in a situation such as a long-distance conference call with multiple speakers in varying environments, for example. Echo cancellation, in at least one embodiment, may be based on similarity measurement, statistical determination of echo cancellation parameters from historical values, frequency domain operation, double talk detection, packet loss detection, signal detection, and noise subtraction.

Abstract: The disclosure provides a method and an apparatus for detecting a voice activity in an input audio signal composed of frames. A noise characteristic of the input signal is determined based on a received frame of the input audio signal. A voice activity detection (VAD) parameter is derived based on the noise characteristic of the input audio signal using an adaptive function. The derived VAD parameter is compared with a threshold value to provide a voice activity detection decision. The input audio signal is processed according to the voice activity detection decision.

Abstract: A tone detector is disclosed that is realizable in digital embodiment on a single integrated circuit die and does not require external components, such as a discrete capacitor. An input connects to a comparator, which in turn connects to one or more edge detectors and a flip flop. The edge detector outputs a pulse responsive to a detected edge. A counter is reset by the pulses from the edge detectors thereby preventing the counter from reaching a maximum value, which would otherwise be output from the counter and provided to a flip flop to clock in the comparator output at the D input to the flip flop. In operation, the comparator generates a rail to rail signal responsive to a received tone, which in turn is clocked through the flip flop as a logic high output indicating presence of a tone.

Abstract: A system that incorporates teachings of the present disclosure may include, for example, one or more components for receiving a call request at a first server from a first end user device, transmitting the call request from a first server to an intermediate server for establishing a voice or video call over an IP multimedia subsystem between the first end user device and a second end user device via a second server, and routing the voice or video call request from the first server to the second server using a second IP address when an undesired condition is determined to be associated with the IP multimedia subsystem. The second IP address can be obtained from a group of IP addresses stored in a memory of the first server and the group of IP addresses can be associated with other servers. Other embodiments are disclosed.

Abstract: A call directing system receives an incoming call from a caller. The caller is prompted to speak, thus enabling a prosody analyzer to generate an analysis of a prosody of the caller's voice. This analysis provides a basis for generating a caller profile that describes caller preferences of the caller. Based on the analysis of the prosody of the caller's voice and the generated caller profile, the call is directed to a particular call recipient.

Abstract: A signaling system is disclosed. The system includes a transmitter comprising an encoder to encode a data signal such that the encoded data signal has a balanced number of logical 1s and 0s. The system also includes a receiver having a decoder to decode the encoded data signal, and a link. The link is coupled between the transmitter and the receiver to route the encoded data signal. The link comprises three or more conductive lines that are routed along a path in parallel between the encoder and the decoder, and wherein the link comprises segments, each segment comprising a routing change to reorder proximity of at least one pair of lines relative to any adjacent segment, with a sufficient number of segments such that each line has each of the other lines of the link as a nearest neighbor over at least a portion of the path.

Abstract: A tone detector is disclosed that is realizable in digital embodiment on a single integrated circuit die and does not require external components, such as a discrete capacitor. An input connects to a comparator, which in turn connects to one or more edge detectors and a flip flop. The edge detector outputs a pulse responsive to a detected edge. A counter is reset by the pulses from the edge detectors thereby preventing the counter from reaching a maximum value, which would otherwise be output from the counter and provided to a flip flop to clock in the comparator output at the D input to the flip flop. In operation, the comparator generates a rail to rail signal responsive to a received tone, which in turn is clocked through the flip flop as a logic high output indicating presence of a tone.

Abstract: The present invention relates to methods and systems for minimizing alien crosstalk between connectors. Specifically, the methods and systems relate to isolation and compensation techniques for minimizing alien crosstalk between connectors for use with high-speed data cabling. A frame can be configured to receive a number of connectors. Shield structures may be positioned to isolate at least a subset of the connectors from one another. The connectors can be positioned to move at least a subset of the connectors away from alignment with a common plane. A signal compensator may be configured to adjust a data signal to compensate for alien crosstalk. The connectors are configured to efficiently and accurately propagate high-speed data signals by, among other functions, minimizing alien crosstalk.

Abstract: A call directing system receives an incoming call from a caller. The caller is prompted to speak, thus enabling a prosody analyzer to generate an analysis of a prosody of the caller's voice. This analysis provides a basis for generating a caller profile that describes caller preferences of the caller. Based on the analysis of the prosody of the caller's voice and the generated caller profile, the call is directed to a particular call recipient.

Abstract: A tone signal received in a telecommunications network is bandpass filtered based on an identified type of the received tone signal, providing a bandpass filtered tone signal. The tone signal is also bandstop filtered based on the identified type of the received tone signal, providing a bandstop filtered tone signal. The bandpass filtered tone signal and the bandstop filtered tone signal are analysed and the received tone signal is quantified based on the analysis result.

Abstract: In one embodiment, a method can include: when performing a park operation: (i) receiving a first message from a park call phone; (ii) providing a second message having a park code to the park call phone; and (iii) receiving a third message having call information from the park call phone; and when performing a retrieve operation: (iv) receiving a fourth message having the park code from a retrieve call phone; (v) providing a fifth message having the call information to the retrieve call phone; (vi) forwarding a sixth message having the park code from the retrieve call phone to a parked device; (vii) forwarding a seventh message from the parked device to the retrieve call phone; and (viii) receiving an eighth message having the park code from the retrieve call phone.

Abstract: A DTMF signal processing apparatus of the present invention comprises a data divider unit, a DTMF signal component analyzer unit, a weighting processing unit, a buffer, a DTMF signal erasure determination unit, and a DTMF signal erasure processing unit. The data divider unit divides speech data into a plurality of divided speech data, and the DTMF signal component analyzer unit analyzes whether or not the divided speech data has a DTMF signal component. The weighting processing unit applies a weighting value to divided speech data analyzed at this time and stores the resultant speech data in the buffer, and also applies a weighting value to past divided speech data previously stored in the buffer when the result of the analysis indicates that the analyzed divided speech data has the DTMF signal component. The DTMF signal erasure determination unit determines based on the weighting value whether or not to erase the divided speech data stored in the buffer.

Abstract: Described herein is a technique for reducing the effects of crosstalk between adjacent signal lines of a data path. The data path is formed by multiple signal lines arranged adjacent each other and traversing multiple segments. The signal lines are transposed between segments in a manner that is chosen to reduce differences in interline couplings between different pairs of the signal lines. The interline coupling of a pair of signal lines is represented as a function of coupling terms. A coupling term corresponds to each segment of a pair of signal lines, and is a function of the distance between the signal lines over that segment. Prior to transmitting a digital signal over the data path, the digital signal is encoded to reduce variations over time in a collective signal level of the digital signal.

Abstract: A method, system and apparatus for reducing dataflow disruption when increases in path maximum transmission unit (PMTU) are being detected are provided. Conventionally, when increases in PMTU are being detected, an increased-in-size data packet is sent by a transmitting host. If an intermediate link is unable to process the data packet through, it drops the data packet and sends an error message to the transmitting host. The transmitting host has to reduce the size of the data packet and resend it. The procedure outlines in this disclosure obviates the need to resend the data packet. It does so by transmitting an old data packet that has been increased in size. Hence, there is no need to resend the data packet if it is dropped.

Abstract: An apparatus, system, and method for telephone signal collection, processing, and analysis is provided. The apparatus includes a portable field unit for capturing telephone signal data including (a) a connection interface for connecting to the telephone line and receiving telephone signal data; (b) a signal recorder for digitally capturing the telephone signal data over a specified period by a plurality of simultaneous sampling methods, wherein a plurality of data streams result from each sampling method; (c) data storage for storing the data streams as a data file; and (d) an output port for transmitting the data file to a remote system. The sampling methods comprise distinct sampling rates and/or distinct input sensitivity levels. The apparatus includes in an alternate embodiment a voice recorder for recording a voice file, wherein the voice file is associated with the data file. This will enable the technician to provide input of his observations of the telephone signal.

Abstract: Disclosed is an integrated tone detection processor for discriminating between tone and voice signals and determining the tones. The tone detection processor performs Automatic Gain Control (AGC) to normalize the power of the tone or voice signal. Further, the energy of the tone or voice signals are determined at specific frequencies utilizing a Goertzel Filter process. The tone detection processor determines whether or not a tone is present, and if a tone exists, determines the type of tone. Based upon determining the two maximum energy levels of the Goertzel filtered tone, whether the tone is a single tone, dual tone, silence, or other (e.g. speech) can be discriminated. The tone can then be identified by a user-defined dictionary of tones. Based upon various ON and OFF cadence checks in combination with the use of TONE ON and TONE OFF counters, tones can be declared.

Abstract: A method and apparatus for detecting a signaling tone within a communications channel, such as a telephone line, is presented. The energy level of the communications channel signal is measured at the frequency of the signaling tone and at one or more frequencies that are integer divisors of the signaling tone (i.e. at one or more subharmonics). The subharmonic energies are evaluated in determining whether the signaling tone is present. The technique can be implemented using a plurality of resonators and a comparator.

Abstract: In an emergency notification and rescue request system having an answer detector for detecting a call receiving party's answer to a telephone call, the answer detector comprises a level detector for outputting a first signal when a level of an audio signal on a telephone line is greater than a prescribed level and otherwise outputting a second signal which is different from the first signal, wherein an interval between adjacent prescribed state changes in the output signal from the level detector is computed every time the prescribed state change is detected, and the current and previous intervals are compared so that if such an event that the current and previous intervals do not substantially correspond to each other has occurred prescribed times consecutively, it is determined that no regular signals such as the busy signal or ring signal exist on the telephone line, i.e., the call receiving party has answered the telephone call.

Abstract: Apparatus and methods for adaptively, reliably, and accurately measuring the duration of an alerting or other signaling tone. A tone detector in accordance with the principles of the present invention accurately measures a duration of a tone burst (i.e., tone pulse) using a long frame length DFT and a short frame length DFT. Based on information obtained by the short frame length DFT, the long (i.e., standard) frame length DFT may be interrupted and reset to begin a new frame of data synchronized with the start of the next short frame corresponding to the processing by the short frame length DFT to minimize noise in the processing by the long frame length DFT and to greatly improve the resolution of the measurement of the duration of a particular tone to correspond to the lengths of the short frames used by the short frame length DFT.

Abstract: A communication terminal comprising a receiving path and a sending path linked to the communication line, the receiving path being connected to a system able to detect, during a communication, an alert signal (e.g., a CAS signal) sent by a station (e.g., the local telephone exchange station or central office) upon any new call or connection request for the terminal and able to allow reception by the terminal, e.g., of data related to each new caller, by interrupting temporarily the presently existing communication, the terminal including a detector (7) able to detect any activity or any signal transmission, and/or the absence of the same, in the sending path (3) of the terminal (1) and to deliver a corresponding information signal (IS) to the signal detection system (5), the latter adjusting its detection method or mode or selecting a corresponding detection subsystem accordingly.

Abstract: A telephone set, where the state of an own-telephone-set is on-hook, the off-hook state of a parallel-telephone-set is detected by comparing telephone-line voltage value with a first standard voltage value. In the state where the own-telephone-set is off-hook, the off-hook state of the parallel-telephone-set is detected by the following method, i.e., the telephone-line voltage value is measured in detail and a second standard voltage value for judging the off-hook state of the parallel-telephone-set is determined and the voltage value is stored in place of the first standard voltage value while the own-telephone-set is off-hooked, then if the line voltage value exceeds the second standard voltage value, the parallel-telephone-set is regarded as off-hook.

Abstract: A surge reset circuit used to automatically reset a surge protection circuit within a telephone line interface circuit. A sensor (e.g., a zenering device) advantageously generates a voltage level representing current draining out of the transient voltage suppression portion of the surge protection circuit after a transient energy surge. If the current is draining, which may indicate a potential lockup situation, feedback circuitry generates a reset feedback signal that is provided to a cutoff device. The cutoff device, in response to receiving the feedback signal, disables a current limiter portion of the surge protection circuit to reduce the current draining from the transient voltage suppression portion. As the current level is reduced, the transient voltage suppression portion will reset when the current drops below a threshold value or is substantially eliminated. As a result, the feedback circuitry removes the reset feedback signal from the cutoff device to re-enable the current limiter.

Abstract: Detection of alerting or other tone signals in a composite signal are reliably detected with an improved false tone detect suppression technique which employs the use of a moving window analysis on a subset of frequencies detected within an underlying speech signal performed during each of a plurality of data frames. The composite signal is transformed during each data frame using a discrete Fourier transform comprising the so-called Goertzel Algorithm. All of the Goertzels relating to the tone detection are analyzed during each data frame, but only a subset of the Goertzels relating to the sweeping harmonic analysis of the underlying speech signal are analyzed during any particular data frame.

Abstract: Apparatus and methods for adaptively, reliably, and accurately measuring the frequency of an alerting or other signaling tone. A tone detector measures the frequency of a tone (continuous or pulsed) with a high degree of accuracy (e.g., to within 1 part in 10,000). Two lower order or smaller DFTs measure the energy level in two separate but intersecting frequency ranges. A simple ratio is determined from the relative energy measured from discrete Fourier transforms relating to each of the two separate frequency ranges, and an actual measured frequency is determined.

Abstract: A telephone with an automatic muting circuit that mutes the microphone in response to detection of speaking pauses of the local user to reduce noise due to acoustical feedback that are detected by comparing the waveforms from the microphone and the audio speaker after being amplified by amplifiers with automatic gain control circuits to the same level with speech subtractor. During pauses when the signals compared are substantially the same, then automatic muting circuit is activated by the speech subtractor to mute the microphone but when the local user speaks into the microphone the subtraction of the waveforms does not yield a null result and the microphone is unmuted.

Abstract: A method and apparatus for verifying that a CAS (Customer Premises Equipment Alerting Signal) detect signal, which indicates the presence of a CAS on a transmission channel, is a valid detect signal, includes structure and process steps for suppressing a signal originating at a local telephone in response to the CAS detect signal. A determination is made whether the CAS detect signal remains in a predetermined condition at a predetermined time after the local telephone signal is suppressed. A validity signal is output when the CAS detect signal remains in the predetermined condition at the predetermined time. False detections are thereby minimized by ignoring detections caused by a signal originating at the local telephone.

Abstract: A system and method for modifying voice signals to avoid triggering tone detectors in a telephone network. The system includes a voice processor coupled between the handset and station set of a conventional telephone unit. The voice processor modifies the voice signal to prevent triggering DTMF tone detectors.

Abstract: A power converter forming part of a telephone subscriber line interface circuit selectively generates a ringing signal waveform, for supply to the line, or a supply voltage for a driver circuit for providing loop current on the line. First and second fixed frequency PWM signals are used to control the power converter, and hence the voltage which it generates, for the ringing signal waveform and the supply voltage respectively. The first PWM signal enables the power converter to provide a high power level needed for ringing signals. The second PWM signal has a lower frequency, and hence results in lower switching losses and power dissipation, and a lower duty cycle, suitable for the lower power level needed for the supply voltage, and is conveniently produced by masking pulses of the first PWM signal. The frequency of the second PWM signal is greater than 270 kHz, to avoid transmitting spectral energy to the line below this frequency.

Abstract: A communication system, and related method, for reliably transmitting DTMF code signals through a low-bit rate channel employing a VSELP speech compression algorithm. The system adds relatively low-level noise to the analog DTMF signals before encoding by the compression algorithm. By adding the low-level noise to the DTMF signals, the tones associated with the DTMF signals can be reliably detected on the receiving end.

Abstract: An input digital sample containing voice and DTMF signal is used as an input to an adaptive filter. The output signal of the adaptive filter is used to generate a residue from the difference between the input sample and the output signal. The residue signal is used to update the coefficients of the adaptive filter. Upon the convergence of the adaptive filter, the residue signal comprises voice reduced DTMF signal while the adaptive filter's output signal contains the detected DTMF signal.

Abstract: A method and device for detecting a customer premises equipment alerting signal (CAS) tone in a telephone system which has both caller identification and call waiting. The present invention distinguishes between valid CAS tones transmitted by the telephone company central office and invalid CAS tones which are caused by music or a human voice. The present invention samples detected tone during a plurality of sample periods. The change of frequency, as well as the direction of frequency change, between successive sample periods is calculated. The number of transitions in the direction of frequency change are counted and used to determine whether the detected tone corresponds to a valid CAS tone or an invalid CAS tone.

Abstract: In a transmission system, a tone is transmitted by a transmitter to a receiver via a transmission channel. In the receiver, a tone detector is used to detect the presence of a signalling tone. In order to improve a reliability of the tone detector when the arrival of the signalling tone is unknown, a number of correlators having mutually displaced measuring periods are used. More than two correlators are used in order to reduce the measuring period, also resulting in an improved reliability of the tone detection.

Abstract: In a transmission system, a signalling tone is transmitted by a transmitter to a receiver via a transmission channel. In the receiver, a tone detector detects the presence of a signalling tone. In order to improve a reliability of the tone detector, a number of correlating elements which determine a correlation value between an input signal and a reference signal are used. Absolute output signals of the correlating elements are added by an adder to derive a combined correlation signal to be used for detection.

Abstract: A telephone apparatus can receive information about a third party sent from a telephone exchange with a call waiting service when a call is issued from the third party during a conversation using the telephone apparatus. Upon issuance of the call from the third party during the conversation over the telephone apparatus, the exchange sends out an alert signal, in response to which the telephone apparatus sends back an acknowledge signal. The telephone apparatus also mutes a speech transmitter circuit in order to protect calling party information such as a telephone number sent from the exchange in succession from being destroyed by a speech signal. The calling party information sent from the exchange can be received without fail and can subsequently be displayed.

Abstract: In a telecommunications system 10 in which remote terminals 20A through 20G are exchanging information with Interactive Information Response Unit (IIRU) 30 via communications in diverse formats and/or media, a system and method for identifying said formats and/or media so that responsive to said identification, appropriate resources 33 may be allocated to the communications to enable IIRU 30 and remote terminals 20A through 20G to exchange information intelligibly.

Abstract: The total energy of a signal that has entered from a telephone line is detected before the signal is applied to an AGC circuit. After the signal is applied to the AGC circuit, the energy of a DTMF signal, which comprises one frequency from four high group frequencies and one frequency from four low group frequencies, is detected through a band-elimination filter having a short energy-propagation delay time with respect to pass-band frequencies. By detecting the DTMF signal based upon the detected energies, detection of the DTMF signal can be performed in a short period of time and erroneous detection can be suppressed.

Abstract: An information signal detection device and an information signal detection method are disclosed with which it is possible to transmit information by means of a DTMF signal using a channel that is prone to distortion, such as a high-efficiency channel. The information signal detection device comprises a filter bank component and a logic component. The filter bank component receives via a channel an information signal in which are superposed sine waves of two different frequencies selected from a specific number of frequencies sent by the sender, and generates binary information patterns showing in binary information whether the signal components of the various frequencies of this information signal are present. The logic component recognizes the end of the signal region when the binary information pattern generated in the binary information pattern generation step is containing no binary information showing that a signal component is present continuously for a specific period of time.

Abstract: A method of detecting voice in an audio signal comprises the steps of determining an average peak value representing an envelope of the audio signal, determining a running instance of audio signal standard deviation, which corresponds to one of a number of overlapping time intervals, and updating a power density function (PDF) by adding instances of noise to the PDF if the average peak of the audio signal exceeds the current level of the audio signal by a certain amount and if the current standard deviation value fails below a threshold for a predetermined time interval. A noise floor is located based on the mean value of the PDF, and, if the audio signal sustains a power level exceeding the noise floor, voice activity is determined to be present in the audio signal.

Abstract: A system for detecting the presence, in an audio channel which may contain voice signals, of a tone at a frequency selectable from among a plurality of selectable frequencies. A bandpass filter receives the audio input; its center frequency is selectable in response to a control signal. A first detector circuit provides from the output of the bandpass filter an output signal when a tone containing substantial energy at about said center frequency is input to the filter. A band-stop filter having the same center frequency as the bandpass filter also receives the input audio; a second detector circuit demodulates and thresholds the band-stop filter output.

Abstract: A system and method for recognizing a voice signal on a telecommunications path upon initiation of an outbound telephone call or receipt of an inbound call by an automated calling system. Power magnitudes of a received signal on the telecommunications path are measured over selected increments of time. Calculations are made to measure the variation in power measured with respect to the selected increments of time, and these calculations are then compared to a threshold value for determining whether or not the received signal is a voice signal.

Abstract: A method and apparatus for verifying that a CAS (Customer Premises Equipment Alerting Signal) detect signal, which indicates the presence of a CAS on a transmission channel, is a valid detect signal, includes structure and process steps for suppressing a signal originating at a local telephone in response to the CAS detect signal. A determination is made whether the CAS detect signal remains in a predetermined condition at a predetermined time after the local telephone signal is suppressed. A validity signal is output when the CAS detect signal remains in the predetermined condition at the predetermined time. False detections are thereby minimized by ignoring detections caused by a signal originating at the local telephone.

Abstract: In a voice processing system, a voice detection algorithm is disclosed wherein the sensitivity is adjusted based upon whether or not tones (e.g.; DTMF) are presently being received. In an additional embodiment, the adjustment is made only when the voice processing system is in a predetermined mode.

Abstract: This tone detector which is suited for dual tone multifrequency (DTMF) digit detection, separates the pairs of tones into a low frequency and high frequency group in which the other frequency group is attenuated by filtering. The amplitude of the low and high frequency groups serve as separate variable threshold reference signals. Individual tones are bandpass filtered with their corresponding outputs converted into an amplitude signal. The amplitude of the individually bandpass filtered tone signals are compared against the corresponding low or high frequency variable threshold signal for a determination of whether a tone is present. Preferably, additional validation steps are applied to the preliminary tone decisions to provide improved digit detection accuracy, especially in the presence of other signals such as speech.

Abstract: A method for detecting the presence of voice, single tone and dual tone signals compares a ratio r which equals the square of the maximum value of the received signal (A max) during a sampling period divided by a measure E of the energy to three different thresholds. If r is less than the first threshold (3), a single tone is indicated. If it is greater than the first but less than the second, the received signal level is compared to the third threshold (-43 dB) and receipt of a voice signal is indicted if the level of the received signal exceeds this threshold.

Abstract: A method of detecting voice in an audio signal comprises the steps of determining an average peak value representing an envelope of the audio signal, determining a running instance of audio signal standard deviation, which corresponds to one of a number of overlapping time intervals, and updating a power density function (PDF) by adding instances of noise to the PDF if the average peak of the audio signal exceeds the current level of the audio signal by a certain amount and if the current standard deviation value falls below a threshold for a predetermined time interval. A noise floor is located based on the mean value of the PDF, and, if the audio signal sustains a power level exceeding the noise floor, voice activity is determined to be present in the audio signal.

Abstract: In methods and apparatus for estimating the frequency response of telecommunications channels, signals carried on the channels are divided into pluralities of signal segments of limited duration, unvoiced signal segments are identified, and spectral components of the unvoiced signal segments are measured. The measured spectral components may be integrated over time to derive time-averaged measurements of the spectral components. The time-averaged measurements may be compared to corresponding components of expected unvoiced signal spectra to improve the accuracy of the frequency response estimates. Filters may be inserted in the channels to adjust the frequency responses, the filters having filter characteristics selected in response to the measured spectral components. The methods and apparatus can be used to compensate for poor bass frequency response in voice channels.