Abstract:
Analysis by synthesis calculates a difference by subtracting (130) synthesized speech from input speech. The synthesized speech is formed by exciting long and short term filters (124, 126) with excitation vectors from a codebook store (114) which is searched by codebook generation (120). A weighting filter (132) is applied to the difference signal and the weighted difference is used to calculate an energy measure (134) which is used to control the codebook search (140). The weighting filter is an Rth-order filter controlled with calculated coefficients. The method for calculating coefficients models the frequency response of L Pth-order filters by a single Rth-order filter, where the order R
Abstract:
A method for protecting information bits wherein input data bits, at least some of which are to be protected, are sorted based upon information determined from a subset of the input data bits. An error control coding technique is applied to at least some of the sorted bits. In the preferred embodiment, an input data stream of voice coder bits is separated into arrays of bits. A first array (302) comprises voice coder bits needing error protection, with the bits arranged in order of importance determined by voicing mode. The second array (303) comprises bits that will not be error protected. The bits from the first array are provided to the input of an encoder (304), then the encoded bits are combined (305) with the bits from the second array (303) to form a bit stream.
Abstract:
A method for protecting information bits wherein input data bits, at least some of which are to be protected, are sorted based upon information determined from a subset of the input data bits. An error control coding technique is applied to at least some of the sorted bits. In the preferred embodiment, an input data stream of voice coder bits is separated into arrays of bits. A first array (302) comprises voice coder bits needing error protection, with the bits arranged in order of importance determined by voicing mode. The second array (303) comprises bits that will not be error protected. The bits from the first array are provided to the input of an encoder (304), then the encoded bits are combined (305) with the bits from the second array (303) to form a bit stream.
Abstract:
Procédé permettant de protéger des bits d'information selon lequel des bits de données d'entrée, dont au moins certains doivent être protégés, sont triés en fonction des informations déterminées provenant d'un sous-ensemble des bits de données d'entrée. On applique une technique de codage de vérification des erreurs à au moins un certain nombre des bits triés. Dans le mode d'exécution préféré, un train de données d'entrée de bits de vocodeur est séparé en deux réseaux de bits. Un premier réseau (302) comprend des bits de vocodeur ayant besoin d'être protégés contre les erreurs, ces bits étant disposés suivant un ordre d'importance qui est déterminé par le mode vocal. Le deuxième réseau (303) comprend des bits qui ne seront pas protégés contre les erreurs. Les bits issus du premier réseau sont fournis à l'entrée d'un codeur (304) et les bits codés sont ensuite combinés (305) avec les bits issus du deuxième réseau (303) pour former un train de bits.
Abstract:
A Vector-Sum Excited Linear Predictive Coding (VSELP) speech coder (200) provides improved quality and reduced complexity over a typical speech coder. VSELP uses a codebook (201) which has a predefined structure such that the computations required for the codebook search process can be significantly reduced. This VSELP speech coder uses single or multisegment vector quantizer of the reflection coefficients based on a Fixed-Point-Lattice-Technique (FLAT). Additionally, this speech coder uses a pre-quantizer to reduce the vector codebook search complexity and a high-resolution scalar quantizer to reduce the amount of memory needed to store the reflection coefficient vector codebooks. Resulting in a high quality speech coder with reduced computations and storage requirements.