排序方式: 共有4条查询结果,搜索用时 15 毫秒
1
1.
2.
3.
在非特定人孤立词语音识别系统中,模板库的容量成为限制识别范围和性能的重要原因,过大的模板库在成本和实时性方面都不利于语音识别系统的大规模使用。提出了一种用于大幅压缩语音识别模板库容量的方法,适用于非特定人孤立词识别系统。该算法借鉴生态学上各物种对专一环境和普通环境的适应程度上的折衷,通过类似不同生物间争夺食物的演化过程选出最具代表性的模板组成模板库。模拟结果表明,该算法通过去除近似雷同的模板以及在某些情况下用一个大模板取代数个模板,能在不显著影响识别率的前提下,明显减少所需模板库的容量。 相似文献
4.
Hereafter, we present a new approach dealing to cope with the harmful effects of noise on speech recognition systems (SRS). This approach is oriented to hardware redundancy and it is essentially a modification of the classic Recovery Blocks scheme. When compared to conventional approaches using Fast Fourier Transform (FFT) and Hamming Code, the primary benefit of such a technique is to improve system performance when operating in real (i.e., noisy) environments. The second advantage is related to the considerably low complexity and reduced area overhead required for implementation. We implemented three full versions of the proposed algorithm: one running of a PC microcomputer, and a second one slightly modified to run on a TMS-320C67 Texas DSP microprocessor module. Both of them were described in the C language. Finally, a last implementation was prototyped on a HW-SW development environment based on the same Texas microprocessor and on the FLEX10K20 FPGA Altera Component. 相似文献
1