Apple Inc.

Charles Pascal Clark

Harvey D Thornburg

Arvindh Krishnaswamy

Sarmad Aziz Malik

An acoustic-scene interpretation apparatus can have a transducer configured to convert an acoustic signal to a corresponding electrical signal. A feature extractor can receive a sequence of frames representing the electrical signal and extract a plurality of acoustic features corresponding to each frame. An acoustic-scene classifier can be configured to determine a most-likely acoustic state for each frame in the sequence of frames in correspondence with the respective plurality of acoustic features corresponding to the frame and a selected probability distribution of duration of an acoustic state for each of one or more classes of acoustic scenes. Each respective probability distribution of duration can correspond to a selected class of acoustic scenes. The correspondence between acoustic state and probability distribution of duration can be learned from training data corresponding to each of a plurality of classes of acoustic scenes. Related methods also are disclosed.

Acoustic scene interpretation systems and related methods

Systems and methods for speech recognition system having a speech processor that is trained to recognize speech by considering (1) a raw microphone signal that includes an echo signal and (2) different types of echo information signals from an echo cancellation system (and optionally different types of ambient noise suppression signals from a noise suppressor). The different types of echo information signals may include those used for echo cancelation and those having echo information. The speech recognition system may convert the raw microphone signal and different types of echo information signals (and optional noise suppression signals) into spectral features in the form of a vector, and a concatenator to combine the feature vectors into a total vector (for a period of time) that is used to train the speech processor, and during use of the speech processor to recognize speech.

Robust speech recognition in the presence of echo and noise using multiple signals for discrimination

Method of detecting voice activity starts with by generating probabilistic models that respectively model features of speech dynamically over time. Probabilistic models may model each feature dependent on a past feature and a current state. Features of speech may include a nonstationary signal presence feature, a periodicity feature, and a sparsity feature. Noise suppressor may then perform noise suppression on an acoustic signal to generate a nonstationary signal presence signal and a noise suppressed acoustic signal. An LPC module may then perform residual analysis on the noise suppressed data signal to generate a periodicity signal and a sparsity signal. Inference generator receives the probabilistic models and receives, in real-time, nonstationary signal presence signal, periodicity signal, and sparsity signal. Inference generator may then generate in real time an estimate of voice activity based on the probabilistic models, nonstationary signal presence signal, periodicity signal, and sparsity signal. Other embodiments are also described.

Growing community of inventors

Mountain View, CA, United States of America

Charles Pascal Clark