The present invention extends to methods, systems, for diagnosing coronary artery disease (CAD) in patients by using their voice signal comprising receiving voice signal data indicative of speech from the patient.

An automated infant cry analyzer with high accuracy to detect important acoustic features of cry is provided. The system's accuracy was rigorously tested and was compared to ground truth manual coding. The resulting methods and systems are applied to infant developmental disorders.

A time-varying pitch of a signal may be estimated by processing a sequence of frames of the speech signal. An estimated fractional chirp rate may be computed for each frame of the sequence of frames, and the estimated fractional chirp rates may be used to compute a pitch template for the sequence, where the pitch template indicates the time-varying pitch of the signal subject to a scale factor. A first pitch estimate for each frame of the sequence of frames may be computed by computing a scale factor and multiplying the pitch template by the scale factor. A second pitch estimate may be computed from the first pitch estimate by identifying peaks in the frequency representations using the first pitch estimates and fitting a parametric function to the peaks.

Audio generation method, server and storage medium are provided. The method includes obtaining a comparison audio, and performing a theme extraction on the comparison audio to obtain a comparison note sequence, the comparison note sequence comprising comparison note positions, comparison note pitches, and a comparison note duration; obtaining an original audio matching with the comparison audio via audio retrieval, and obtaining an original note sequence corresponding to the original audio by performing a theme extraction on the original audio, the original note sequence comprising original note positions, original note pitches, and an original note duration; calculating theme distances between fragments of the comparison audio and fragments of the original audio according to the comparison note sequence and the original note sequence; and generating an audio by capturing a fragment that is of the original audio and that satisfies the smallest theme distance.

The present document discloses a method and apparatus for compensating for a lost frame in a transform domain, comprising: calculating frequency-domain coefficients of a current lost frame using frequency-domain coefficients of one or more frames prior to the current lost frame, and performing frequency-time transform to obtain an initially compensated signal; and performing waveform adjustment, to obtain a compensated signal. Alternatively, extrapolation is performed for all or part of frequency points of the current lost frame using phases and amplitudes of corresponding frequency points of a plurality of previous frames to obtain phases and amplitudes of the corresponding frequency points of the current lost frame, to obtain frequency-domain coefficients of the corresponding frequency points, and frequency-time transform is performed to obtain a compensated signal. The above methods can be selected through a judgment algorithm to compensate for the current lost frame, thereby achieving a better compensation effect.

A speech processing method for estimating a pitch frequency includes: executing a conversion process that includes calculating a spectrum from a plurality of frames included in an input signal; executing a determination process that includes determining a speech-like frame from the plurality of frames based on characteristics of the spectrum of the frame; executing a learning process that includes specifying a fundamental sound based on a plurality of local maximum values included in the spectrum of the speech frame and learning a learning value based on a magnitude of the fundamental sound; and executing a detection process of detecting a pitch frequency of the frame based on the spectrum of the frame and the learning value.

A pitch information generation device includes a first envelope generator configured to generate, with regard to a first sound range, a first envelope that attenuates at a first rate of change from a detected value corresponding to a peak in the sound signal, a second envelope generator configured to generate, with regard to a second sound range, which includes a sound range of higher frequency than the first sound range, a second envelope that attenuates from a detected value corresponding to a peak in the sound signal at a second rate of change. The second rate of change is greater than the first rate of change. A pitch information identifier is configured to identify the pitch information based on the first envelope and the second envelope.

Disclosed herein is a system for identifying and diagnosing sounds of infected wild birds and poultry. The system includes: a multi channel audio analysis device (MCAAD); and a sound collection unit connected to the MCAAD via a wired or wireless connection. The sound collection unit collects sounds of birds via a plurality of microphones, and information about the collected sounds is transmitted to the MCAAD via a relay.

Machine learning technology can analyze in real-time the data from a call between a person and a customer service representative. Based on this analysis, a server can determine a sentiment score that describes a sentiment expressed by the person or the customer service representative. If the server determines that the sentiment score is less than or equal to a pre-determined value, the server can inform the customer service representative's manager so that the manager can take further action to help the person and/or the customer service representative.

A conversation meter comprises a memory storage comprising instructions and one or more processors in communication with the memory storage. The one or more processors execute the instructions to perform: accessing audio data representing a conversation among a plurality of people; analyzing the audio data to associate one or more portions of the audio data with each person of the plurality of people; analyzing the portions of the audio data to determine one or more conversation metrics for each person of the plurality of people; and causing presentation of at least one of the determined conversation metrics.