Linear predictive coding

About: Linear predictive coding is a research topic. Over the lifetime, 6565 publications have been published within this topic receiving 142991 citations. The topic is also known as: Linear predictive coding, LPC.

Prototype waveform speech coding with interpolation of pitch, pitch-period waveforms, and synthesis filter

Abstract: A speech coding system providing reconstructed voiced speech with a smoothly evolving pitch-cycle waveform. A speech signal is represented by isolating and coding prototype waveforms. Each prototype waveform is an exemplary pitch-cycle of voiced speech. A coded prototype waveform is transmitted at regular intervals to a receiver which synthesizes (or reconstructs) an estimate of the original speech segment based on the prototypes. The estimate of the original speech signal is provided by a prototype interpolation process which provides a smooth time-evolution of pitch-cycle waveforms in the reconstructed speech. Illustratively, a frame of original speech is coded by first filtering the frame with a linear predictive filter. Next a pitch-cycle of the filtered original is identified and extracted as a prototype waveform. The prototype waveform is then represented as a set of Fourier series (frequency domain) coefficients. The pitch-period and Fourier coefficients of the prototype, as well as the parameters of the linear predictive filter, are used to represent a frame of original speech. These parameters are coded by vector and scalar quantization and communicated over a channel to a receiver which uses information representing two consecutive frames to reconstruct the earlier of the two frames based on a continuous prototype waveform interpolation process. Waveform interpolation may be combined with conventional CELP techniques for coding unvoiced portions of the original speech signal.

Data selection for speech recognition

Abstract: This paper presents a strategy for efficiently selecting informative data from large corpora of transcribed speech. We propose to choose data uniformly according to the distribution of some target speech unit (phoneme, word, character, etc). In our experiment, in contrast to the common belief that "there is no data like more data", we found it possible to select a highly informative subset of data that produces recognition performance comparable to a system that makes use of a much larger amount of data. At the same time, our selection process is efficient and fast.

A speaker-independent digit-recognition system

Abstract: This paper describes an implementation of a speaker-independent digit-recognition system The digit classification scheme is based on segmenting the unknown word into three regions and then making categorical judgments as to which of six broad acoustic classes each segment falls into The measurements made on the speech waveform include energy, zero crossings, two-pole linear predictive coding analysis, and normalized error of the linear predictive coding analysis A formal evaluation of the systems showed an error rate of 27 percent for a carefully controlled recording environment and a 56 percent error rate for on-line recordings in a noisy computer room

High-quality speech at low bit rates: Multi-pulse and stochastically excited linear predictive coders

Abstract: We will present in this paper some recent developments in low bit rate speech coding research. Two new speech coding algorithms - multi-pulse LPC and stochastic coding (code-excited linear prediction) - have been proposed recently to achieve high quality speech at bit rates below 10 kbits/sec. Both of these algorithms use a linear filter to synthesize speech at the receiver but they differ in the methods used to generate the excitation for the linear filter. The multi-pulse model assumes that the excitation can be represented with sufficient accuracy as a sequence of pulses (typically 4 to 8 pulses every 5 msec). In stochastic coders, the excitation is selected from a random codebook of white Gaussian sequences. The optimum excitation in both these coders is chosen to minimize a subjective error criterion based on properties of human auditory perception. Although these coding algorithms are complex requiring over 10 million multiply-add operations per second, new fast digital signal processor chips offer the possibility of their real-time implementation.

Data driven example based continuous speech recognition.

Abstract: The dominant acoustic modeling methodology based on Hidden Markov Models is known to have certain weaknesses Partial solutions to these flaws have been presented, but the fundamental problem remains: compression of the data to a compact HMM discards useful information such as time dependencies and speaker information In this paper, we look at pure example based recognition as a solution to this problem By replacing the HMM with the underlying examples, all information in the training data is retained We show how information about speaker and environment can be used, introducing a new interpretation of adaptation The basis for the recognizer is the wellknown DTW algorithm, which has often been used for small tasks However, large vocabulary speech recognition introduces new demands, resulting in an explosion of the search space We show how this problem can be tackled using a data driven approach which selects appropriate speech examples as candidates for DTW-alignment