Andrew P. Dejaco

Method for determining speech encoding rate in a variable rate vocoder

In a variable rate vocoder a method for determining a higher encoding rate of a set of encoding rates for unvoiced speech. The method is accomplished by generating an encoding rate indication based upon a first characteristic of an audio signal, determining a second characteristic of the audio signal, and modifying the encoding rate indication when the second characteristic of the audio signal is representative of unvoiced speech to provide a modified encoding rate indication corresponding to a higher encoding rate of the set of encoding rates.

Multimode variable bit rate speech coding: an efficient paradigm for high-quality low-rate representation of speech signal

The speech signal consists of a time-varying ensemble of different types of segments with distinct characteristics, which require different degrees of coding resolution in order to retain an overall high voice quality. A fixed-rate coder can capture such time-varying characteristics only if it operates at a high enough bit rate. At a low bit rate, a fixed-rate coder will not be able to capture all of these various segments well and will fail to render high voice quality. A multimode variable bit rate (VBR) coder uses an arsenal of modes, operating at different bit rates. These modes are designed to represent these different speech segments optimally with the right amount of coding resolution. Thus, a multimode VBR codec adapts the coding mechanism to the input speech and delivers high quality at low (average) rates. This paper presents the essential framework and the unique advantages of a multimode VBR codec and suggests algorithms for the different modes.

A robust speech recognition system embedded in CDMA cellular phone chipsets

PureVoice™ VR is a speech recognition system characterized by noise robustness performance, small footprint, and limited computing requirement. The system was embedded into Qualcomms Code Division Multiple Access (CDMA) Mobile Station Modem (MSM) chipsets. The PureVoice™ VR engine enabled many manufactures of CDMA cellular phones to add highly accurate voice-activated dialing features to their handsets and hands-free car kits without significant increase in hardware cost. Users of these cellular phones could make calls by simply speaking names or phone numbers. The current embedded system has the capacity to recognize 100 speaker-dependent (SD) nametags and user defined commands, or 30 speaker-independent (SI) digits and commands in realtime. It also works in a hybrid mode, in which an input utterance can be either SD namtags or SI commands. PureVoice™ VR has successfully supported the voice-activated dialing function for millions of CDMA handsets sold in the world market.