Bruce Alan Fette

Variable frame rate, fixed bit rate vocoding method

A method of operating a vocoder so that a variable frame rate results while maintaining a constant bit rate is disclosed. A base frame rate remains constant. However, spectral change is measured to determine an appropriate subframe rate within the base frame. Only one of a plurality of separate vector quantization processes is selected in response to spectral change for performance of a vector quantization operation on LPC spectrum coefficients. Each of the plurality of vector quantization operations utilizes its own codebook that contains a different quantity of reference patterns from that contained in the other codebooks. Each vector quantization operation produces a reference pattern descriptor code that contains a different number of bits from that produced by the other vector quantization processes. Vector quantization operations producing smaller, less spectrally accurate outputs are selected when more subframes are included with a base frame, and vector quantization operations producing larger, more spectrally accurate outputs are selected when fewer subframes are included within a base frame.

Digital voice processing system

A multiple rate voice processing system incorporating a complete linear predictive coding algorithm wherein the algorithm is partitioned among a plurality of integrated circuit chips so that all communications between chips occur at low data rates.

Human voice analyzing apparatus

The analyzing apparatus includes a ten stage, all-zero lattice digital filter formed on a single semiconductor chip. A partial correlation coefficient is derived in each stage of the lattice filter in improved coefficient circuitry and the analyzer provides the ten partial correlation coefficients, for a best sample from a plurality of samples, along with the amplitude (R.M.S.), and the residual energy, or the excitation. The correlator uses the products FB, F2 and B2, of the residual F and B signals.

Experiments with a high quality, low complexity 4800 bps residual excited LPC (RELP) vocoder

The authors present an approach for achieving high quality speech coding at 4800 bps using a residual excited LPC (RELP) coder. Numerous experiments were performed to identify and code the significant parameters in the spectral, pitch, and residual models. Major principles for high quality residual coding are discussed.<<ETX>>

Windowing functions for the average magnitude difference function pitch extractor

This paper describes the fundamental properties of the Average Magnitude Difference Function, (AMDF) when used for pitch extraction on human speech, such as the ability to work in the absence of the fundamentals, and responses to pitch and formant resonances. Requirements for the algorithms correct functioning are discussed, including reasonable bounds on window size and requirements on a nonlinear post processor. Finally, three experiments are described: 1) comparison of rectangular window versus exponential window AMDF, 2) comparison of the accuracy of the indicated pitch versus window length, and 3) effects of inverse filtering on the AMDF.

Method and apparatus for implementing vector quantization of speech parameters

A vocoder (10) includes a digital signal processor (18) and a vector quantizer (28) implemented in hardware external to the digital signal processor (18). The vector quantizer (28) includes an analysis codebook (32) having a multitude of scalar quantized code vectors (60). A frame of digitized speech is translated into speech parameters, which are then scalar quantized to form a source vector. This scalar quantized source vector is then vector quantized in the vector quantizer (28) prior to transmission into a communication channel (22). A separate synthesis codebook (30) is used in decompressing vocoded speech received from the communication channel (22).

High Quality 2400 Bps Vocoder Research

Global trends toward complex, high-capacity telecommunications emphasize a growing need for high quality speech coding techniques that require less bandwidth. Telecommunications networks of the future, such as Motorolas IRIDIUM, will demand very high fidelity voice communications at the lowest possible bit rates. Inadequate vocal system models of current 2400 bps vocoder technologies often fail to reproduce the full character and resonance of the original speech, and are thus unacceptable for systems requiring high-quality voice communications. Motorola research in the high-quality 2400 bps arena has resulted in innovative solutions to high quality, low-bandwidth voice communication requirements.

SDR strategies for information warfare and assurance

As software defined radios (SDR) proliferate and the capability and functionality of radios expand, the opportunities for attack by either side increases. Modern networks are evolving into a combination of wired, wireless, and Internet components with attacks possible on any component against any other component. Understanding those attacks to identify vulnerabilities and formulate defensive approaches is the first step in a comprehensive system design. Furthermore, a complete understanding of adversarial vulnerabilities enables development of offensive strategies that leverage the power of network centric warfare (NCW). Computational capabilities of emerging SDRs provide the means to coordinate attacks against the adversary. Further, these capabilities will enable new methods for overcoming classical attacks against the terrestrial, UAV, and Satcom information systems. Part of the protection strategy for these systems will require that classical information assurance (IA) techniques be distributed to various SDR nodes within the infrastructure.

CipherVOX: scalable low-complexity speaker verification

Biometrics is gaining strong support for access control in the industry. It is not uncommon for individual users to be faced with a half-dozen or more passwords and personal identification numbers (PINs) controlling access to the systems required for them to do their job. The ubiquity of passwords actually relaxes system security since many users tend to use the same password across all applications, or collect the various passwords in a single location (perhaps a password protected spreadsheet, or a piece of paper in a desk drawer). The latter case is of extreme concern since security around the collection is much more easily compromised than that of the system. The use of biometrics not only recovers the ability to secure sensitive systems and data, but also does so in a user-friendly manner. In order to use biometrics successfully in server-based environments, several key concerns must be addressed. First, the authentication strategy must maintain acceptable levels of security. Second, the user community must accept the chosen biometric (unless there is a captive audience). The last major consideration is the scalability of the solution. In this paper we introduce CipherVOX, a speaker verification access control solution for server and standalone computing environments. We discuss the use of our polynomial-based classifier that combines high-accuracy and low-complexity via discriminative techniques, and give performance results for both a proprietary performance database and the standard YOHO database. We also review the challenges in designing for user acceptance, including the design of the speaker verification user interface, as well as the application programming interface (API).

A manpack portable LPC 10 vocoder

A manpack portable LPC-10 Vocoder has been developed which makes substantial size and power performance improvements over existing LPC Vocoders. This is accomplished through the use of distributed processing using the latest VLSI digital signal processing technology and advanced microcomputer technology. First the LPC-10 standard algorithm was partitioned into high performance digital signal processing tasks that may stand alone as VLSI devices. The remaining LPC-10 algorithmic components are partitioned by the data and process flow graphs into meaningful multi-purpose stand alone single chip computers, resulting in a vocoder that uses 3 VLSI, and 3 LSI components. The digital signal processing algorithms are partitioned as follows: LPC Analysis IC, LPC Synthesis IC, AMDF Pitch Extraction IC. The data flow processes are partitioned into microcomputers as follows: Transmit Pitch and Voicing in processor #1, Transmit AGC in processor #2, and Parameter Quantization and Serialization in processor #3; in the receive mode sync acquisition and maintenance and parameter deserialization in processor #3, error correction and dequantization in processor #2, and interpolation rule implementation in processor #1. The data flow processors partitions were greatly affected by the use of single chip computers. These computers have a very limited RAM and ROM space causing the partition to be dependent on program size. The use of single chip computers minimizes external hardware necessary for the vocoder implementation.