Code-excited linear prediction

About: Code-excited linear prediction is a research topic. Over the lifetime, 2025 publications have been published within this topic receiving 28633 citations. The topic is also known as: CELP.

Quality enhancement of narrowband CELP-coded speech via wideband harmonic re-synthesis

Abstract: Results for improving the quality of narrowband CELP-coded speech by enhancing the pitch periodicity and by regenerating the high-band components of speech spectra are reported. Multiband excitation (MBE) analysis is applied to enhance the pitch periodicity by re-synthesizing the speech signal using a harmonic synthesizer. The high-band magnitude spectra are regenerated by matching to low-band spectra using a trained wideband spectral codebook. Information about the voiced/unvoiced (V/UV) excitation in the high-band is derived from a training procedure and recovered by using the matched low-band index. Simulation results indicate that the quality of the wideband enhanced speech is significantly improved over the narrowband CELP-coded speech.

Comparison of two speech content authentication approaches

Abstract: Speech content authentication, which is also called speech content integrity or tamper detection, protects the integrity of speech contents instead of the bitstream itself. In this work, two major approaches for flexible speech authentication are presented and compared. The first scheme is based on content feature extraction that is integrated with CELP speech coders to minimize the total computational cost. Speech features relevant to semantic meaning are extracted, encrypted and attached as the header information. The second method embeds fragile watermarks into the speech signal in the frequency domain. If the speech signal is tampered, the secret watermark sequence is also modified. The receiver detects the fragile watermark from received data and compares it to the original secret sequence. These two approaches are compared in terms of computational complexity, false detection rate, and tolerance to mis-synchronization and content preserving operations. It is shown that each approach has its own merits and shortcomings.

Method for processing speech signals as block floating point numbers in a CELP-based coder using a fixed point processor

Abstract: A method of encoding speech using a fixed-point processor. The method treats the signal as floating point, while operating on each sample of the signal as fixed point. The disclosed method achieves precision similar to that of conventional floating point and may be rapidly executed on a fixed point processor.

LSP analysis and processing for speech coders

Abstract: Linear prediction parameters within CELP coders are commonly represented by line spectral pairs (LSP), giving stable filters and efficient coding. However, LSP manipulation can also alter the frequencies of the represented signals. The authors use computationally efficient LSP manipulation to enhance the intelligibility of speech degraded by acoustic interference.

Performance Analysis of a Decoder-Based Time Scaling Algorithm for Variable Jitter Buffering of Speech Over Packet Networks

Abstract: This paper describes how a CELP speech decoder such as the VMR-WB decoder can be modified in order to deliver decoded speech frames of variable length, a feature that is required for adaptive jitter buffering. This method is shown to be successful for playout delay adaptation in terms of both subjective quality and reactivity. Moreover, it requires almost no additional complexity provided some clever limitations are imposed on time scaling.