3785 results about "Background noise" patented technology

A method for recognizing an audio sample locates an audio file that most closely matches the audio sample from a database indexing a large set of original recordings. Each indexed audio file is represented in the database index by a set of landmark timepoints and associated fingerprints. Landmarks occur at reproducible locations within the file, while fingerprints represent features of the signal at or near the landmark timepoints. To perform recognition, landmarks and fingerprints are computed for the unknown sample and used to retrieve matching fingerprints from the database. For each file containing matching fingerprints, the landmarks are compared with landmarks of the sample at which the same fingerprints were computed. If a large number of corresponding landmarks are linearly related, i.e., if equivalent fingerprints of the sample and retrieved file have the same time evolution, then the file is identified with the sample. The method can be used for any type of sound or music, and is particularly effective for audio signals subject to linear and nonlinear distortion such as background noise, compression artifacts, or transmission dropouts. The sample can be identified in a time proportional to the logarithm of the number of entries in the database; given sufficient computational power, recognition can be performed in nearly real time as the sound is being sampled.

Audio frames are classified as either speech, non-transient background noise, or transient noise events. Probabilities of speech or transient noise event, or other metrics may be calculated to indicate confidence in classification. Frames classified as speech or noise events are not used in updating models (e.g., spectral subtraction noise estimates, silence model, background energy estimates, signal-to-noise ratio) of non-transient background noise. Frame classification affects acceptance / rejection of recognition hypothesis. Classifications and other audio related information may be determined by circuitry in a headset, and sent (e.g., wirelessly) to a separate processor-based recognition device.

Techniques for canceling echo and suppressing noise using an array microphone and signal processing. In one system, at least two microphones form an array microphone and provide at least two microphone input signals. Each input signal may be processed by an echo canceller unit to provide a corresponding intermediate signal having some echo removed. An echo cancellation control unit receives the intermediate signals and derives a first gain used for echo cancellation. A noise suppression control unit provides at least one control signal used for noise suppression based on background noise detected in the intermediate signals. An echo cancellation and noise suppression unit derives a second gain based on the control signal(s), cancels echo in a designated intermediate signal based on the first gain, and suppresses noise in this intermediate signal based on the second gain. The signal processing may be performed in the frequency domain.

An eyewear with communications capability to pair with a Bluetooth enabled device and equipped with noise cancellation software. The earpieces are connected to a retraction wheel by a flexible cord. The cord winds about the wheel to retract the earpiece and unwinds to extend the earpiece to its operative position. A retractable button may be pressed to urge the retraction wheel to rotate to wind the cord about it under spring force.

Methods, apparatus, and products are disclosed for adjusting a speech engine for a mobile computing device based on background noise, the mobile computing device operatively coupled to a microphone, that include: sampling, through the microphone, background noise for a plurality of operating environments in which the mobile computing device operates; generating, for each operating environment, a noise model in dependence upon the sampled background noise for that operating environment; and configuring the speech engine for the mobile computing device with the noise model for the operating environment in which the mobile computing device currently operates.

A sensor unit that includes at least one sensor configured to measure an ambient condition is described. The controller can be configured to receive instructions, to report a notice level when the controller determines that data measured by the at least one sensor fails a report threshold test corresponding to a report threshold value. The controller can also be configured to obtain a plurality of calibration measurements from the at least one sensor during a calibration period and to adjust the threshold based on the calibration measurements. The controller can be configured to compute a first threshold level corresponding to background noise and a second threshold level corresponding to sensor noise, and to compute the report threshold value from the second threshold. In one embodiment, the sensor unit adjusts one or more of the thresholds based on ambient temperature.

A method of providing speech transcription performance indication includes receiving, at a user device data representing text transcribed from an audio stream by an ASR system, and data representing a metric associated with the audio stream; displaying, via the user device, said text; and via the user device, providing, in user-perceptible form, an indicator of said metric. Another method includes displaying, by a user device, text transcribed from an audio stream by an ASR system; and via the user device, providing, in user-perceptible form, an indicator of a level of background noise of the audio stream. Another method includes receiving data representing an audio stream; converting said data representing an audio stream to text via an ASR system; determining a metric associated with the audio stream; transmitting data representing said text to a user device; and transmitting data representing said metric to the user device.

A flow cytometer (20) capable of utilizing an imaging system including a sensor (88) to determine various properties of the flow cytometer (20). Importantly, the imaging allows for determination of a drop delay time (154). Other characteristics that can be determined include at least the width of the stream, pressure of the stream, effect of charged droplets on other droplets (44), trajectory of the stream, resonant frequency, wavelengths, change in position of droplet break-off point (150), and others. An automatic warning system (180) can be used to alert an operator of an anomaly during setup or during normal operation. Furthermore, a mechanical interrupter can be used to shield a sort result from contamination by an incorrectly functioning stream. The cytometer (20) also allows for the disablement of the sort, particularly the charging and deflection systems. In addition, a more accurate system for detecting the speed of the stream can be used. The imaging system including a sensor (88) can allow for the removal of background noise and the monitoring of a change in a stream characteristic.

Methods, systems, and products for testing a grammar used in speech recognition for reliability in a plurality of operating environments having different background noise that include: receiving recorded background noise for each of the plurality of operating environments; generating a test speech utterance for recognition by a speech recognition engine using a grammar; mixing the test speech utterance with each recorded background noise, resulting in a plurality of mixed test speech utterances, each mixed test speech utterance having different background noise; performing, for each of the mixed test speech utterances, speech recognition using the grammar and the mixed test speech utterance, resulting in speech recognition results for each of the mixed test speech utterances; and evaluating, for each recorded background noise, speech recognition reliability of the grammar in dependence upon the speech recognition results for the mixed test speech utterance having that recorded background noise.

An apparatus for detecting sound includes a plurality of microphones, a sound inspecting unit, a direction estimating unit, a background noise removing unit, and an alerting unit. The microphones are used to collect sounds around a user. The sound inspecting unit is used to calculate the feature values of a background noise within a preset time interval, and to determine if a latest collected sound satisfies a preset condition. When the preset condition is satisfied, the direction estimating unit is used to estimate the occurrence direction of the latest collected sound, and to determine if the occurrence direction is within a preset range behind the user. When the preset range is satisfied, the background noise removing unit is used to remove the background noise in the latest collected sound so as to obtain a detected sound. The alerting unit is used to inform the user of the detected sound via an alert message. A method for detecting sound is also disclosed.

A physiological sensing stethoscope suitable for use in high-noise environments is disclosed. The stethoscope is designed to be substantially matched to the mechanical impedance of monitored physiological activity and substantially mismatched to the mechanical impedance of air-coupled acoustic activity. One embodiment of the stethoscope utilizes a passive acoustic system. Another embodiment utilizes an active Doppler system. The passive and active systems can be combined in one stethoscope enabling switching from a passive mode to an active mode suitable for use in very high-noise environments. The stethoscope is suitable for use in environments having an ambient background noise of 100 dBA and higher. The passive includes a head having a housing, a flexural disc mounted with the housing, and an electromechanical stack positioned between the housing and the flexural disc in contact with the skin of a patient. The active system detects Doppler shifts using a high-frequency transmitter and receiver.

A hearing compensation system comprises a plurality of bandpass filters having an input connected to an input transducer and each bandpass filter having an output connected to the input of one of a plurality of multiplicative automatic gain control (MAGC) circuits whose outputs are summed together and connected to the input of an output transducer. The MAGC circuits attenuate acoustic signals having a constant background level without the loss of speech intelligibility. The identification of the background noise portion of the acoustic signal is made by the constancy of the envelope of the input signal in each of the several frequency bands. The background noise that will be suppressed includes multi-talker speech babble, fan noise, feedback whistle, florescent light hum, and white noise. For use in the consumer electronics field background acoustic noise may be sensed and used to adjust gain in the various MAGC circuits so as to improve a user's listening experience, whether the user is hearing impaired or not.

A method and neurostimulator for providing therapy to a patient is provided. In one technique, electrical background energy is conveyed to a first tissue region of the patient in accordance with stochastic parameter, thereby modulating the excitability of the first tissue region, and electrical stimulation energy is conveyed to the first tissue region when its excitability is modulated. In one example, the stimulation energy may be conveyed to a second tissue region of the patient, thereby therapeutically stimulating the second tissue region. In this case, the excitability of the first tissue region is decreased, thereby reducing any adverse effect that the conveyed stimulation energy has on the first tissue region. As another example, the conveyed stimulation energy stimulates the first tissue region, in which case, the excitability of the first tissue region may be increased, thereby enhancing the stimulation of the first tissue region by the conveyed stimulation energy.

Natural-quality synthetic noise will replace background acoustic noise during speech gaps and will achieve a better representation of the excitation signal in a noise-synthesis model by classifying the type of acoustic environment noise into one or more of a plurality of noise classes. The noise class information is used to synthesize background noise that sounds similar to the actual background noise during speech transmission. In some embodiments, the noise class information is derived by the transmitter and transmitted to the receiver which selects corresponding excitation vectors and filters them using a synthesis filter to construct the synthetic noise. In other embodiments, the receiver itself classifies the background noise present in hangover frames and uses the class information as before to generate the synthetic noise. The improvement in the quality of synthesized noise during speech gaps helps to preserve noise continuity between talk spurts and speech pauses, and enhances the perceived quality of a conversation.

In an information processing system having an equipment controller for enabling the digitization of complex work processes conducted during testing and / or operation of machinery and equipment, a local wireless communications network is implemented using at least one fixed point wireless communications access station and one or more voice-responsive wireless mobile computing / communications devices that are capable of accurate vocal command recognition in noisy industrial environments. The computing / communications device determines spectral characteristics of ambient non-speech background noises and provides active noise cancellation through the use of a directional microphone and an adaptive noise tracking and subtraction process. The device is carried by a user and is responsive to one or more vocal utterances of the user for communicating data to the information processing system and / or generating operational control commands to provide to the equipment controller for controlling the machinery or equipment.

In one aspect thereof the invention provides a method for noise suppression of a speech signal that includes, for a speech signal having a frequency domain representation dividable into a plurality of frequency bins, determining a value of a scaling gain for at least some of said frequency bins and calculating smoothed scaling gain values. Calculating smoothed scaling gain values includes, for the at least some of the frequency bins, combining a currently determined value of the scaling gain and a previously determined value of the smoothed scaling gain. In another aspect a method partitions the plurality of frequency bins into a first set of contiguous frequency bins and a second set of contiguous frequency bins having a boundary frequency there between, where the boundary frequency differentiates between noise suppression techniques, and changes a value of the boundary frequency as a function of the spectral content of the speech signal.

A remote sensor for use as a handheld, mobile or stand-alone sensor has first (12) and second (16) optical paths, light collecting optics, a sample filter (10) assembly positioned in a first optical Path (12), a reference filter (14) assembly positioned in a second optical path (16), a detector assembly to detect the filtered light r other radiation, and a detector output comparison device such as BRD to minimize the effects of common background noise components, differences in light or other radiation source power, and absorption or emission by interfering species.

Embodiments of methods and devices for classifying background noise contained in an audio signal are disclosed. In one embodiment, the device includes a module for extracting from the audio signal a background noise signal, termed the noise signal. Also included is a second that calculates a first parameter, termed the temporal indicator. The temporal indicator relates to the temporal evolution of the noise signal. The second module also calculates a second parameter, termed the frequency indicator. The frequency indicator relates to the frequency spectrum of the noise signal. Finally, the device includes a third module that classifies the background noise by selecting, as a function of the calculated values of the temporal indicator and of the frequency indicator, a class of background noise from among a predefined set of classes of background noise.

Improved pilot signal sequences which facilitate multiple channel quality measurements, e.g., through the use of different signal pilot transmission power levels, are described. In various implementations the transmitted pilot sequences facilitate determining the contribution of interference from other sectors of a cell using the same tones, e.g., in a synchronized manner, as the sector in which the pilot signal measurements are being made. To measure noise contributions from neighboring sectors a sector NULL pilot, e.g., a pilot with zero power, is transmitted in an adjacent sector at the same time a pilot signal with a pre-selected, and therefore known, non-zero power is transmitted in the sector where the received pilot signal measurement is made. To facilitate background noise measurements, a cell NULL is supported in some embodiments. In the case of a cell NULL, all sectors of a cell transmit a Null pilot, on a tone that is used to measure background noise. Since no power is transmitted in the cell on the tone during the measurement, any measured signal on the tone is attributable to noise, e.g., background noise which may include inter-cell interference.

A mobile audio device (for example, a cellular telephone, personal digital audio player, or MP3 player) performs Audio Dynamic Range Control (ADRC) and Automatic Volume Control (AVC) to increase the volume of sound emitted from a speaker of the mobile audio device so that faint passages of the audio will be more audible. This amplification of faint passages occurs without overly amplifying other louder passages, and without substantial distortion due to clipping. Multi-Microphone Active Noise Cancellation (MMANC) functionality is, for example, used to remove background noise from audio information picked up on microphones of the mobile audio device. The noise-canceled audio may then be communicated from the device. The MMANC functionality generates a noise reference signal as an intermediate signal. The intermediate signal is conditioned and then used as a reference by the AVC process. The gain applied during the AVC process is a function of the noise reference signal.

A UAV is provided to cancel background noise from audio data collected by the UAV. The UAV is provided with one or more background microphones in a proximity of one or more background noise-producing components. The UAV is also provided with one or more audio source collecting microphones. The audio data collected by the background microphones may be used to reduce or cancel interfering background noise from the audio signal detected by the audio source collecting microphone. The target audio may be captured or recorded with little or no background noise.

Audio signal processing relating to the measurement and control of the perceived sound loudness and / or the perceived spectral balance of an audio signal is useful, for example, in one or more of: loudness-compensating volume control, automatic gain control, dynamic range control (including, for example, limiters, compressors, expanders, etc.), dynamic equalization, and compensating for background noise interference in an audio playback environment. In various embodiments, modification parameters are derived for modifying the audio signal in order to reduce the difference between its specific loudness and a target specific loudness.

A method for recognizing an audio sample locates an audio file that most closely matches the audio sample from a database indexing a large set of original recordings. Each indexed audio file is represented in the database index by a set of landmark timepoints and associated fingerprints. Landmarks occur at reproducible locations within the file, while fingerprints represent features of the signal at or near the landmark timepoints. To perform recognition, landmarks and fingerprints are computed for the unknown sample and used to retrieve matching fingerprints from the database. For each file containing matching fingerprints, the landmarks are compared with landmarks of the sample at which the same fingerprints were computed. If a large number of corresponding landmarks are linearly related, i.e., if equivalent fingerprints of the sample and retrieved file have the same time evolution, then the file is identified with the sample. The method can be used for any type of sound or music, and is particularly effective for audio signals subject to linear and nonlinear distortion such as background noise, compression artifacts, or transmission dropouts. The sample can be identified in a time proportional to the logarithm of the number of entries in the database; given sufficient computational power, recognition can be performed in nearly real time as the sound is being sampled.

A method of processing a far-end signal and a near-end signal to produce a final signal, the far-end signal containing speech, the near-end signal containing speech and background noise. The method includes: determining an amplification gain based upon the near-end signal; removing a portion of the background noise from the near-end signal to create a noise-reduced near-end signal; combining the far-end signal with the noise-reduced near-end signal to create a combined signal; and amplifying the combined signal by the amplification gain to create the final signal.

An apparatus for enhancing audibility of a far-end speech signal from a far-end user to a near-end user in a telephony system includes a near-end background noise signal level estimator (26) at the near-end user. A near-end speech signal level estimator (24) is also provided at the near-end user. A gain control logic (22) determines a gain (G) for amplification of the far-end speech signal based on both estimated speech and background noise signal levels.