搜索资源列表
recnet-1.3.tar
- recnet is a complete speech recognition system for the DARPA TIMIT and Resource Management tasks. It uses recurrent networks to estimate phone probabilities and Markov models to find the most probable sequence of phones or words.-recnet is a co
ffmpeg-r10408-dll
- Movie2Mobile, written in Python, is a tool for transcoding your favorite movie into the 3gp format, used by most mobile phones. For encoding, MPlayer, MEncoder and FFMpeg is used, to offer a wide range of input formats, including DVD, AVI/MPEG and (S
Great_Outdoors_by_sandals82.zi
- 一种简单有效的基于动态时变语音识别源码 对于大多数研究者来说,寻找能够匹配二重时间序列信号的最佳途径是很重要的,因为它有许多重要的应用需求.DTW是实现这项工作的显著技术,尤其在语音识别技术领域,在这里一个测试信号被按照参照模板拉伸或压缩, ,Searching for the best path that matches two time-series signals is the main task for many researchers, because of its importa
521834
- 提供软电话的处理方法，对声卡的输入输出进行控制，提供参考-Soft phones provide ways of addressing the issue of the sound card input and output control, to provide reference
julius-4.3.1.tar
- Julius 是一种高性能，两通大词汇量连续语音识别（LVCSR）语音相关的研究和开发的解码器软件。基于字的N-gram和上下文相关的HMM模型，它可以进行几乎实时实时解码目前大多数电脑在60K字听写任务。完全纳入，如树的N-gram词汇，保，跨词的上下文依赖处理，包围梁搜索，高斯修剪，高斯的选择，除了搜索效率等各大搜索技术，它也是模块化小心从模型结构独立，如共享状态triphones的和并列混合模型与任意数量的混合物，州或手机，支持各种HMM的类型。采用标准格式，以配合HTK的，债务工具中央结