基于DSP音信号线性与非线性量化相互转化的新方法.docVIP

摘　　要 随着通信、计算机网络等技术的飞速发展,语音压缩编码技术得到了快速发展和广泛应用。尤其是最近20年，语音压缩编码技术在移动通信、卫星通信、多媒体技术以及IP电话通信中得到普遍应用，起着举足轻重的作用。人们相互交流的信息量也在不断地急剧增加，庞大的语音信号数据给存储和传输带来了巨大的的压力，使得信道资源变得愈加宝贵。因此，语音压缩和语音编码技术显得越来越重要。 本课题是基于DSP语音信号线性与非线性量化相互转化的新方法，通过DSP将采集到的语音信号进行压缩算法的处理。最后通过外设输出压缩后的语音信号。最终实现语音信号的采集、压缩与回放。本论文根据系统的功能需求，完成了该系统的算法研究，软硬件的设计。设计出了A律编解码的软件流程框图，在以TMS320VC5502为处理器的硬件开发平台上实现了语音信号的A律压缩解压算法，并给出了压缩程序流程图。 Abstract With the communications, computer networks of rapid development, voice compression coding technology has been rapid of development and wide of application. Especially in the last 20 years, speech coding technology was widely application in the mobile communications, satellite communications, multimedia and IP telephony technology, it plays a pivotal role. People mutually exchanging information is increasing dramatically, huge voice signal data to the storage and transmission brought huge pressure, it makes channel resources become more and more valuable. Therefore, speech compress and speech coding technology is becoming more and more important. This topic is based on the DSP of voice compression algorithm design and implementation. The collected voice signal use compression algorithm to treat by DSP. The speech signal after compression is output by external equipments finally. It has realized the speech signal collection、compression and playback finally. According to the systems functional requirements, this papers complete hardware and software design of the system. A law designed a flow chart of the software codec in order to TMS320VC5502 processor hardware development platform for the realization of the speech signal on the A-law compression decompression algorithms, And give the compression process flow diagram. 引 言 语音是人类相互进行交流时使用最多、最自然、最基本也是最重要的信息载体。语音的产生是一个复杂的过程，包括心理和生理等方面的一系列动作。由于其特殊的作用，人们历来十分重视对语音信号和语音通信的研究。 近十几年来语音技术在人们实际需要的推动下快速的发展起来，语音技术是一个跨学科、涉及面广的综合学科，包括声学、语音学、生理学、心理学、数字信号处理、信息工程、通信理论、电子科学、模式识别、人工智能等众多学科，而且许多对语音数字信号处理有促进作用的学科如神经网路、小波理论、遗传算法、进化算法、模糊理论、混沌理论等也在蓬勃发