搜索资源列表
ostream-Text-To-Speech
- linux 下的 简单语音识别 代码 可以参考下-ostream-Text-To-Speech
linux_test_new_7z
- 包含一些脚本处理的题目,并含有一个自动识别语音的程序。-Contains a number of scr ipts deal with the subject and contains an automatic voice identification procedures.
ENDPOINTDETECT
- 语音识别的端点检测程序,通过能量,过零率MFCC低能量普等检测-Endpoint detection of speech recognition procedures, the adoption of energy, zero-crossing rate of MFCC & P, such as low-energy detection
sphinxbase-0.4.1
- sphinxbase是卡内基梅隆大学著名的sphinx语音识别工程的公用库-sphinxbase Carnegie Mellon University, the famous sphinx speech recognition works Public Library
EmotionalCreature-master
- 模拟生物情感程序,包括情感分析,语音的变速变调,语音的合成和识别。-Simulated bio-affective procedures, including the emotional analysis, a shift of voice tone change, etc.. .
latticelm-0.3
- 语音识别中处理lattice的工具,可以进行后验概率计算、剪枝等处理-Speech recognition processing lattice tools that can be posterior probability calculation, pruning and other processing
openlat-master
- 语音识别中在openfst上进行lattice处理的代码,以WFST理论为基础-Speech Recognition on the lattice in openfst handling code to WFST theory
VCPP
- VC++ 语音识别程序源代码 入门级别参考的书-VC++ speech recognition program source code entry-level reference book
htk-3.3
- HTK是英国剑桥大学开发的一套基于C语言的隐马尔科夫模型工具箱,主要应用于语音识别、语音合成的研究,也被用在其他领域,如字符识别和DNA排序等。HTK是重量级的HMM版本。-Cambridge University HTK is a C-based language developed by the Hidden Markov Model Toolkit, mainly used in speech recognition, speech synthesis research, has also
raspberrySDK
- raspbain系统的科大讯飞语音识别动态库 官方发的是.a静态库 我重新编译成了.so的动态库 方便大家用jni或者其他调用 内置demo 关于java调用的demo 本人不久会上传-raspbain system IFLYTEK Speech Recognition dynamic library The official made the .a static library .so I recompiled became dynamic libraries to facilit
Linux_voice_1.015_559a3710
- 讯飞语音识别合成SDK包,基于Linux,为开发者提供语音识别和语音合成云接口。-IFlyTek synthesized speech recognition SDK package, based on Linux, voice recognition and speech synthesis Cloud API for developers.
lingyun_asr_server.cpp.tar
- 灵云语音识别,完成音频到文字的转换,处理pcm格式-convert audio to text
EndPoint
- 语音端点检测,刚打了飞机的考拉法国大的放大范德萨发(voice endpoint by c++,coiodjogj;agldfhkgk;alfjdo;ashkgdf)
WEGASUN-M6语音识别模块资料包2017-6-19
- 裸机编程,实现led灯的读写,使用TQ210开发版(Baremetal programming)
asr_sample
- 科大讯飞的语音识别案例。基于linux平台实现。(A case of speech recognition of iFlyTEK)
dtw
- 可以用来实现dtw。在孤立词语音识别中,最为简单有效的方法是采用DTW(Dynamic Time Warping,动态时间归整)算法,该算法基于动态规划(DP)的思想,解决了发音长短不一的模板匹配问题,是语音识别中出现较早、较为经典的一种算法,用于孤立词识别。HMM算法在训练阶段需要提供大量的语音数据,通过反复计算才能得到模型参数,而DTW算法的训练中几乎不需要额外的计算。所以在孤立词语音识别中,DTW算法仍然得到广泛的应用。(Can be used to implement DTW)