Audio signal processing

About: Audio signal processing is a research topic. Over the lifetime, 21463 publications have been published within this topic receiving 319597 citations. The topic is also known as: audio processing & Acoustic signal processing.

Multi-channel digital video transmission receiver with improved channel-changing response

Abstract: A multi-channel digital video receiver (e.g. DBS) includes a tuning unit with two tuner modules and a decoder unit with two decoder modules. These tuner and decoder units enable the receiver to simultaneously produce decoded video and audio signals for two separately-tuned channels. The signals for the channel currently selected by the user are tuned and decoded by one set of tuner and decoder modules. The decoded signals for the selected channel are sent to an output stage (e.g. encoder), where the signals and their output are encoded for use on an attached presentation device (e.g. television, audio processor, computer, etc.) Simultaneously, a microcontroller predicts the next channel the user will select and causes the other tuner and decoder modules to begin tuning to and decoding the signals for that predicted next channel. A third set of tuner and decoder modules can be added to enable the receiver to begin tuning to and decoding the signals for another likely next channel. Since the receiver begins tuning to the predicted next channel even before a new channel is requested, an expected channel-change command can be processed more quickly than in a conventional receiver.

Factorial Scaled Hidden Markov Model for polyphonic audio representation and source separation

Abstract: We present a new probabilistic model for polyphonic audio termed Factorial Scaled Hidden Markov Model (FS-HMM), which generalizes several existing models, notably the Gaussian scaled mixture model and the Itakura-Saito Nonnegative Matrix Factorization (NMF) model. We describe two expectation-maximization (EM) algorithms for maximum likelihood estimation, which differ by the choice of complete data set. The second EM algorithm, based on a reduced complete data set and multiplicative updates inspired from NMF methodology, exhibits much faster convergence. We consider the FS-HMM in different configurations for the difficult problem of speech / music separation from a single channel and report satisfying results.

Audio tuning system

Abstract: An audio system installed in a listening space may include a signal processor and a plurality of loudspeakers. The audio system may be tuned with an automated audio tuning system to optimize the sound output of the loudspeakers within the listening space. The automated audio tuning system may provide automated processing to determine at least one of a plurality of settings, such as channel equalization settings, delay settings, gain settings, crossover settings, bass optimization settings and group equalization settings. The settings may be generated by the automated audio tuning system based on an audio response produced by the loudspeakers in the audio system. The automated tuning system may generate simulations of the application of settings to the audio response to optimize tuning.

Portable audio playback device with bass enhancement

Abstract: Wirelessly linking a portable audio device to an external audio system permits the external audio system to generate audible signals that enhance local playback of audio by the portable device. For example, the portable device wirelessly transmits lower frequency components of an audio signal to a subwoofer system for reproduction of bass frequencies extending below the playback capability of the portable device. In this manner, the external audio system provides bass enhancement for the portable audio device. Wireless transmissions between the portable audio device and the external audio system may be, but are not limited to, optical or radio frequency (RF) transmissions. Where RF signaling is used, the wireless link may be based on wireless network links, such as those supported by Bluetooth and 802.11b standards, or based on, for example, dedicated RF interfaces.

Multiple clock rate teleconferencing network

Abstract: A hub-resident data switching apparatus for a multinode teleconferencing network enables plural node sites, operating at differing clock rates, to communicate with each other asynchronously and simultaneously. The switching apparatus includes an internal TDM bus to which plural node interface units, an audio combiner unit and a timing and control unit are coupled. Each node interface unit is associated with a respective node and is operative to receive and forward communication signals with respect to its node at the clock rate of the service used by that node. The hub's TDM bus includes a video/PC file bus and an audio/command bus and operates at a clock rate that is a multiple of the number of nodes in the network times the highest clock frequency of any node in the network divided by the bit width of the TDM bus. Each unit interfaces with the TDM bus by way of a pipeline bus interface unit. The audio combiner unit outputs digitized `combined audio` signal packets onto the audio bus for transmission to its associated node. The output section of the audio combiner unit includes an audio packet input buffer whose contents are processed in dependence upon the occupancy status of each audio packet input buffer in the combiner. As long as there is sufficient audio data stored in one or more buffers to ensure effectively continuous audio processing, audio data is processed and forwarded for transmission. If the audio packet input buffers have very little or no audio data, then processing is delayed.