Study and Development of Time Varying Bit-rate MELP Vocoder

马欣,李文元,刘常澍,张毓忠

DOI: https://doi.org/10.3969/j.issn.1002-8684.2004.05.013

2004-01-01

Abstract:The2Kb/s MELP(Multi Excitation Linear Prediction)speech CODEC has been newly adopted as U.S.Federal standard.In this paper,a varying bit rate MELP CODED model is implemented.This model accomplishes relatively good synthetic speech quality under1.5k～2Kb/s.

What problem does this paper attempt to address?

Study and development of MELP vocoder

Li Wenyuan,Xin Ma,Zhang Yuzhong,Zeng Yifang

DOI: https://doi.org/10.1109/ICOSP.2002.1180125

2002-01-01

Abstract:In mixed-excitation linear prediction (MELP) coding, a new speech synthesis model that approaches actual human pronunciation mechanism is adopted to improve the performance of low bit-rate speech coding. A penetrating analysis of the MELP algorithm is presented in this paper. Improvements in multi-stage vector searching method and in pitch estimation are suggested. A 2.4 kbit/s MELP vocoder has been simulated on PC and realized real time using one single TMS320C5402 DSP.
New Mixed Excitation Linear Prediction Codec at 2.4 kb/s

Jian-ling HU,Sheng XU,Jian CHEN

DOI: https://doi.org/10.3321/j.issn:1006-2467.2000.06.017

2000-01-01

Shanghai Jiaotong Daxue Xuebao/Journal of Shanghai Jiaotong University

Abstract:A new mixed excitation linear prediction (MELP) codec working at very low bit rate was presented. This codec utilizes the advantages of both linear prediction coding (LPC) algorithm and multi-band excitation (MBE) coding algorithm. In order to get natural speech with good quality, some new features were employed for the algorithm and quantization scheme including robust pitch detection algorithm, mixed excitation determination, efficient linear spectrum frequency (LSF) parameter quantization, and new excitation spectrum expression. Informal subjective testing shows that the quality of speech synthesized by a 2.4 kb/s codec using this MELP algorithm is a little better than or same as that synthesized by FS1016 4.8 kb/s CELP.
Design and Description of a 600 bps Speech Coder Based on Melpe

Feng Zou,Ying Guo,Xinfu Chen,Yan Liu

DOI: https://doi.org/10.1109/ITW2.2006.323820

2006-10-01

Abstract:This paper describes a 600 bps speech coder based on the enhanced mixed excitation linear prediction model (MELPe). The algorithm of this speech coder includes features of MELPe, which can obtain high quality synthesized speech and is robust in difficult background noise environments. To reduce the bit rate, we have developed a modified multi-frame joint vector quantization that takes advantage of inherent inter-frame redundancy. The predicted multi-stage vector quantizer (PMSVQ) is designed to quantize the line spectrum frequency (LSF) parameters. Simulation results have proven that an efficient and high quality coding has been achieved at bit rate 600 bps, and the proposed coder is better than the existing 2400 bps LPC10e standard

Computer Science
An Improved 2.4kb/s Mixed Excitation Linear Prediction Vocoder

Ma Xin,Li Wenyuan,Liu Changshu,Zhang Yuzhong

DOI: https://doi.org/10.1007/bf02687915

2005-01-01

Journal of Electronics (China)

Abstract:This letter presents two improvements on 2.4 kb/s Mixed-Excitation Linear Prediction (MELP) vocoder. The one is a new parameter Redzc named energy to differential zerocrossing rate which is used in adaptation of V/UV decision of transitional segments and low energy level speech segments. The other is a multi-path searching method for Multi-Stage Vector Quantization (MSVQ) of line spectral frequency. Subjective tests show that the intelligiblity and naturallity of improved MELP vocoder are preferable to those of the original one.
Realtime robust speech communication based on iterative joint source-channel decoding and demodulation algorithm for MELP vocoder

Tan Peng,Huijuan Cui,Kun Tang

DOI: https://doi.org/10.3772/j.issn.1006-6748.2010.02.001

2010-01-01

Abstract:Realtime speech communications require high efficient compression algorithms to encode speech signals. As the compressed speech parameters are highly sensitive to transmission errors, robust source and channel decoding and demodulation schemes are both important and of practical use. In this paper, an iterative joint source-channel decoding and demodulation algorithm is proposed for mixed excited linear prediction ( MELP) vocoder by both exploiting the residual redundancy and passing soft information throughout the receiver while introducing systematic global iteration process to further enhance the performance. Being fully compatible with existing transmitter structure, the proposed algorithm does not introduce additional bandwidth expansion and transmission delay. Simulations show substantial error correcting performance and synthesized speech quality improvement over conventional separate designed systems in delay and bandwidth constraint channels by using the joint source-channel decoding and demodulation (JSCCM) algorithm.
Research on Low Delay Low Bit Rate Speech Coding Algorithm

赵哲峰,张刚,谢克明,王一平

DOI: https://doi.org/10.3778/j.issn.1002-8331.2008.34.031

2008-01-01

Computer Engineering and Applications Journal

Abstract:The 16 kb/s Low Delay Code-Excited Linear Prediction(LD-CELP) has been used in many environments,however,the high bit rate required in this work have been a handicap in using it for some condition.In this paper,a speech coder using two stages code book search schemes is proposed to improve the performance of LD-CELP and reduce its bit rate.To develop the proposed speech coder with a low bit rate,two sub code books are explored:a backward updated adaptive sub code book and a fix sub code book with algebraic structure.Second,a two stages target vector search scheme is proposed to minimize the mean squared error between the filtered excitation vector and the target vector.As a result,a 10 kb/s dual code book CELP coder is proposed with a delay of 2.5 ms in sampling rate of 8 kHz.It is shown that the proposed speech coder has a comparable speech quality to 16 kb/s G.728 in Segmented Signal-to-Noise Ratio(SSNR) and Perceptual Evaluation of Speech Quality(PESQ).
Research on MBE Algorithm at Bit Rate 800 Bps-2.4 Kbps Vocoder

Huijuan Cui,Kun Tang,Haisheng Zheng

DOI: https://doi.org/10.1109/icct.1998.741013

1998-01-01

Abstract:This paper presents a variable-rate speech coding at 800 bps, 1.2 kbps and 2.4 kbps based on the algorithm of multi-band excitation (MBE). It takes the advantages of MBE algorithm in pitch estimation, UV decision and the estimation of spectral amplitude vector. The main difficulty of at the low bit rate of the MBE algorithm lies in an efficient quantization of the spectral amplitude vector. The technique of linear prediction (LP) is used, and the spectral amplitude vectors are approached by an all-pole model with formants to be enhanced. The techniques of vector quantization (VQ) is adopted to quantize the LP parameters. The split-VQ (SPVQ), and multi-stage VQ (MSVQ) are employed to reduce the storage capacity and computing complexity, and max-min method is used to accelerate the searching of the VQ codebook. The 800 bps, 1.2 kbps and 2.4 kbps vocoder has been implemented in real-time using a single fixed point DSP ADSP2181 chip. The synthesized speech of this vocoder is clear and intelligible even in the presence of background noise at 80 dB. The speech quality of this vocoder has reached the current international level
Research on real time low bit-rate speech coding

Huijuan Cui,Haisheng Zheng,Hao Jiang,Tian Wang,Kun Tang

1997-01-01

Abstract:Low bit rate speech coding is one of the most important and fundamental parts of modern digital communication network and multimedia techniques. This paper presents a high quality variable-rate speech coding at 2.4 kbit/s, 1.2 kbit/s and 800 bit/s based on the algorithm of multi-band excitation (MBE). The techniques of smoothing pitch period by using dynamic programming algorithm, LPC all-pole model to approximate the MBE spectral shape, formant enhancement, split vector quantizer, multi-stage vector quantizer, multilayer perception of neural networks, etc. are adopted to improve this vocoder. It is implemented in realtime by using a single ADSP2181 fixed point digital signal processor. The speech quality of the real-time system is satisfactory.
An Improved MELP Speech Coder

Chaogang Wu,Hongxu Jiang,Bo Li

DOI: https://doi.org/10.1109/ITCS.2009.166

2009-01-01

Abstract:This paper presents several strategies to improve the performance of MELP speech coder. A new approach is given to analyze and synthesize the transition segments which are called the transition frame. This can improve the accuracy of parameter estimation and reduce the noise caused by classing the transition frame simply into U/V frame. A new frame type decision algorithm with dynamic U/V threshold related to the pitch lag is introduced which classifies the speech frame into three types: voiced, unvoiced and transition. The classifier can reduce the U/V decision errors and avoid excessive switching between voiced frame and unvoiced frame. These strategies are incorporated into a speech codec M-MELP. The quality tests show that M-MELP is superior to the MELP coder, especially for high-frequency female speakers.
A DSP-based Low Rate Speech Coding Technology in Short-Wave Communications

CHEN Hua,GUAN Yu-dong,WANG Jian

DOI: https://doi.org/10.3969/j.issn.1009-671x.2007.08.003

2007-01-01

Abstract:A 1.5 Kbps low rate speech coding algorithm applied in short-wave communications is introduced,which takes advantages of the vocoder model of MELP(mixed excitation linear prediction) speech coding algorithm,so that the complexity and bit rate are reduced.The algorithm was run in real time in a hardware system containing DSP chip.Finally the simulation results are presented.
A Low Bit Rate Speech Coding Applied to Phonetic Communications in Satellite Maritime Tracking & Controlling

陈松,胡剑凌,周朝猛

DOI: https://doi.org/10.3969/j.issn.1001-893x.2004.05.028

2004-01-01

Abstract:This paper focuses on how to solve the problem of safeguarding the phonetic communications between the astronaut and the command headquarters in case of emergency in the space mission performed by measure-ship. In order to realize low bit rate speech coding, Mixed Excitation Linear Prediction (MELP) vocoder, a new speech encoding apparatus,is applied to phonetic communications in measure-ship, especially to the amalgamation between the new technique and the former systerm.
A Hybrid Structure Speech Coding Scheme Based on MELPe and LPCNet

Yaoyao Ding,Xinle Yu

DOI: https://doi.org/10.1109/eebda56825.2023.10090529

2023-01-01

Abstract:Although the conventional speech coding algorithms can achieve 1200bps, 600bps or even lower, the synthesized speech quality becomes worse and worse, and the improvement effect is limited. Several speech synthesis models based on deep learning proposed in recent years make it possible to synthesize high-quality speech. This paper integrates LPCNet into the traditional MELPe algorithm to finish the speech synthesis task. Experiments demonstrate that the quality of this scheme is similar to that of MELPe at 1.07kbps, which can be used as a new idea to study the ultra-low rate MELPe.
Improvement of Mixed Excitation Linear Prediction Speech Coding Algorithm and Implementation on DSP System

Xiaochen Wu,Longxiang Guo,Yang,Nana Wu,Haining Lv,Wei Jiang,Yanning Wang,Xiaotian Jiang

DOI: https://doi.org/10.1121/1.4875896

2014-01-01

Abstract:With the rapid development of the communication technology, the high quality of speech communication has become one of the main development trends. In order to solve the quality problem in the processing of communication, mixed excitation linear prediction speech coding algorithm (MELP) because of its higher compression rate and better security performance, is widely used in communication systems. In this paper pre-emphasis is used to improve the quality of the high frequency components in this algorithm. The speech enhancement of spectral subtraction is also applied in this speech coding algorithm to improve the ability of anti-noise. Adaptive difference pulse code improvement is also used in the quantization part of the speech algorithm to obtain the transmission rate that is much lower than before. Furthermore, the algorithm is transplanted into the digital signal processing system in order to validate the feasibility of this algorithm. At last, the speech quality will be proved by PESQ - MOS test and ...
Research on Speech Coding Algorithm at Very Low Bit Rate for Narrow-Band Communication

LIU Bin,TAO Jian-hua,MO Fu-yuan

DOI: https://doi.org/10.3969/j.issn.1003-0530.2013.09.007

2013-01-01

Journal of Signal Processing

Abstract:A speech coding algorithm applied to narrowband transmission at very low bit rate is proposed.Based on the MELP standard working at 2.4kbps,joint vector quantization is applied to linear spectrum pair's parameters,interpolation and nonlinear quantization are applied to pitch,gain parameters are compressed efficiently and auditory perception is considered,speech data could be coded and transmitted smoothly at 0.5kbps The predicative residuals of linear spectrum pair's parameters are also used for vector quantization;it 's a suitable way of improving synthetic speech quality.The experimental results reveal that the proposed algorithm can ensure speech effective transmission at very low bit rate while the intelligibility of received speech is acceptable.
A High Quality MBE-LPC-FE Speech Coder at 2.4 Kbps and 1.2 Kbps.

T Wang,K Tang,CX Feng

DOI: https://doi.org/10.1109/icassp.1996.540327

1996-01-01

Abstract:This paper presents a high quality speech coder based on the multi-band excitation (MBE) model operating at 2.4 kb/s and 1.2 kb/s. The features of this coder mainly focus on two aspects. One is an accurate and reliable pitch estimation and voiced/unvoiced decision algorithm. The other is the efficient quantization of the variable dimension spectral amplitude vector. Besides representing the spectral envelope information using an all-pole model, we also encode error vectors of the spectral amplitude vector at several important positions. Informal listening tests indicate that the speech quality of this coder at 2.4 kb/s is comparable and even superior to the INMARSAT-M IMBE 4.15 kb/s coder. The coder has been implemented in real-time on a single TMS320C31 floating point DSP.
High Quality Harmonic Excitation Linear Predictive Speech Coding at 2 Kb/s

CC Bao,J Lukasiak,C Ritz

DOI: https://doi.org/10.1109/chinsl.2004.1409608

2004-01-01

Abstract:The paper presents a high quality harmonic excitation linear predictive (HE-LPC) speech coder operating at 2 kb/s based on a harmonic excitation model with two bands. The system incorporates novel features such as: combined pitch detection; residual harmonic matching voicing determination; extraction and interpolation of residual harmonic magnitudes. Subjective listening tests indicate that this coder has the same quality as that of the Federal Standard MELP (mixed excitation linear prediction) coder at 2.4 kb/s, whether the training database is from Chinese or English.
Multiband Excitation Coding of Speech at 1.8 ～ 2.4 Kbps

张军,肖自美,韦岗

DOI: https://doi.org/10.3321/j.issn:0371-0025.2002.05.004

2002-01-01

Acta Acustica

Abstract:A new fast pitch-refining algorithm of MBE model, which is based on the feature of frequency domain error function around pitch, and a new scalable quantization scheme of spectral envelope magnitude, which is based on MBE-LPC model and Variable-Dimension Vector Quantization (VDVQ), are proposed. A 1.8 - 2.4 kbps scalable MBE coder is implemented with the new pitch-refining algorithm and the new quantization scheme. Experiment result shows that this MBE coder achieves the same synthesis quality as the original MBE-VDVQ coder in the same bit rate with much less computation and memory .
An Extreme Low Bit Rate Speech Coding Algorithm Around 300bps

Gang Min,Jian Guo,Jibin Yang,Wei Tan,Yanpu Chen

DOI: https://doi.org/10.1109/WCSP.2009.5371427

2009-01-01

Abstract:An extreme low bit rate speech coding algorithm around 300bps is proposed in this paper. The algorithm builds mixed excitation segment coding model by taking advantage of the segment coder and the MELP coder. Variable dimension matrix quantization (VDMQ) and Variable dimension vector quantization (VDVQ) scheme are presented for quantizing LSP and excitation parameters. These quantization schemes achieve acceptable performance at very low bit rate. Also, the codebook storage is reduced dramatically. Informal subjective listening test shows that the reconstructed speech has high intelligibility and moderate naturalness, the PESQ score can achieve 2.02.
Design and Implementation of a Low Bit Rate Programmable Vocoder

荆涛,刘恩涛,王沁

2008-01-01

Abstract:For the safety enhancement of digital speech communication and the utilization improvement of transmission channel bandwidth,a new special purpose processor for speech compression based on low bit-rate SELP algorithm is proposed,which characterizes programmable,high degree of parallelism,reconfigurable and safety.Four-stage recombinant ALU and variable length VLIW instruction set are adopted to optimize the function modules with high complexity and enhance the degree of instruction level parallelism.Simulation results indicate that the average consumption is 200mW under the main frequency 20MHz when the vocoder implements the 0.6kbps SELP algorithm,and the operating complexity of the vocoder is 12.5MIPS which is much lower than that of general purpose DSP under the same state of the art.
Optimization for Algorithm of MELP Speech Codec Based on DSP Platform

LI Yang,WU Yu-cheng

DOI: https://doi.org/10.3969/j.issn.1002-8684.2010.05.010

2010-01-01

Abstract:In order to meet the need of engineering application, the algorithm of the MELP(Mixed-Excitation Linear Prediction) speech codec on the DSP platform needs to be optimized. Based on the data, the code of the MELP speech codec is optimized by using different methods which includes compiler tools, small function inline expansion and intrinsics and so on. Furthermore, by analyzing and comparing the performance of various optimization methods and results, it shows that the tested results meet the practical application with good effects and practical significance.

Study and Development of Time Varying Bit-rate MELP Vocoder

Study and development of MELP vocoder

New Mixed Excitation Linear Prediction Codec at 2.4 kb/s

Design and Description of a 600 bps Speech Coder Based on Melpe

An Improved 2.4kb/s Mixed Excitation Linear Prediction Vocoder

Realtime robust speech communication based on iterative joint source-channel decoding and demodulation algorithm for MELP vocoder

Research on Low Delay Low Bit Rate Speech Coding Algorithm

Research on MBE Algorithm at Bit Rate 800 Bps-2.4 Kbps Vocoder

Research on real time low bit-rate speech coding

An Improved MELP Speech Coder

A DSP-based Low Rate Speech Coding Technology in Short-Wave Communications

A Low Bit Rate Speech Coding Applied to Phonetic Communications in Satellite Maritime Tracking & Controlling

A Hybrid Structure Speech Coding Scheme Based on MELPe and LPCNet

Improvement of Mixed Excitation Linear Prediction Speech Coding Algorithm and Implementation on DSP System

Research on Speech Coding Algorithm at Very Low Bit Rate for Narrow-Band Communication

A High Quality MBE-LPC-FE Speech Coder at 2.4 Kbps and 1.2 Kbps.

High Quality Harmonic Excitation Linear Predictive Speech Coding at 2 Kb/s

Multiband Excitation Coding of Speech at 1.8 ～ 2.4 Kbps

An Extreme Low Bit Rate Speech Coding Algorithm Around 300bps

Design and Implementation of a Low Bit Rate Programmable Vocoder

Optimization for Algorithm of MELP Speech Codec Based on DSP Platform