1-Written by Arthur Chan 2 3Developers of Sphinx Here we list all Sphinx's developes who have 4contributed to Sphinx. Since, in the early stage of Sphinx's 5development we didn't use RCS and CVS. This version only include the 6contributions which can be derived from the source code. The 7description may not be accurate enough. Please kindly correct us if 8there is any descripencies of this file. 9 10There are also countless contributers for Sphinx. It is possibly 11impossible to list them in a single file. Do protest to us if you 12want your name to be somewhere here. We are more than willing to do 13it. 14 15-Arthur Chan. 20040825 16 17Kai Fu Lee - The one who proposed Sphinx. 18 19Hsiao-wen Hon - According to Dr. Kai-Fu Lee's "Sphinx: a speaker 20independent speech recognition system", Hsiao-wen is the implementer 21of Sphinx I. 22 23Fil Alleva - 24Robert Brennan - 25 26Mei-Yuh Hwang - The proposer of the concept of senones. 27 28Xue Dong Huang- Manager of Sphinx 2. A lot of improvement were 29introduced from Sphinx 1 to Sphinx 2. 30 31Ravishankar Mosur - Implementer of Sphinx 2(fbs-6,7,8) and Sphinx 3 32(both flat lexicon decoder and tree lexicon decoder. Many fast 33algorithms in S2 and S3 were designed and implemented by him. 34 35Eric Thayer - Architect of SphinxTrain, he unified a lot of toolkits 36of CMU's researchers and create the backbone of the current 37SphinxTrain. 38 39Rita Singh - Proposer and Implementer of automatic question 40generation. She also implemented the first version of live mode 41decoder and write a comprehensive on-line tutorial for SphinxTrain, 42sphinx2 and sphinx3. Many researchers were benefited by her version 43of modified code. 44 45Bhiksha Raj - Implementer of silence deletion in the alignment in 46s3align and a version of SAT in SphinxTrain. Main designer of Sphinx 474. 48 49Ricky Houghton - He is the author of the backbone of the perl scripts 50for SCHMM training. He also fixed a lot of memory leaking problems in 51Sphinx 3. 52 53Mike Seltzer - He is the author of the current version of wave2feat. 54 55Sam-Joo Doh - He is the author of a speaker adaptation package. The 56current version of adaptation routine only used small portion of its 57package. 58 59Kevin Lenzo - Packaged Sphinx 2 and developed several versions of 60Sphinx2 to make it has open-source quality. He also started to put 61the code into Sourceforge. 62 63Alan W Black - Packaged SphinxTrain's script and code. He makes it work 64for speech synthesis. 65 66Evandro Gouvea - Package Sphinx3.0 and Sphinx 3.x, maintainer of 67Sphinx X including the code, websites, machine queues and networks. 68 69Developers from Sun, including Willie Walker, Paul Lamere and Philip 70Kwok. They first ported s3.3 to s3j, a java version of s3.3. Later, 71they proposed to design and implement Sphinx 4. For Sphinx 4's 72contributors, please consult Sphinx 4's documentation. 73 74Jahanzeb Sherwan - Implementer of phoneme lookahead algorithm in 75Sphinx 3.x 76 77Yitao Sun - Designer and Implementer of Sphinx 3.x live mode APIs. 78 79David Huggins-Daines - Implementer of unsupervised speaker adaptation 80routine. 81 82Scott Silliman - Tester of libutil library. Help us to fence out a 83lot of bugs. 84 85Arthur Chan - The author of this document. :-) Before I left CMU, 86let's make this description empty. :-) 87 88Special Thanks 89 90Carl Quillen - He submitted a patch for allowing three s3.0 tools to 91s3.x. This allows developers of Sphinx to incorporate other tools 92very quickly to s3.x. He also submitted a patch of mean 93transformation. 94 95Joseph Donohue - He gave us a lot of invaluable advices on code 96integrity. We took most of it and was benefited a lot. 97 98Ziad Al Bawab - He implemented the current end-pointing routine. 99 100Rong Zhang - He implemented the C-version of confidence annotation 101routines. 102 103Michael Pust, Jia Pu and Keith Herold - They give us invaluable 104suggestions and advices on how to improve the code quality. We used 105their suggestions to adopt the use of doxygen. 106 107Thanks 108 109Rich Stern - Experts in biaurnal speech processing and he 110manages CMU's robust group. 111 112Alex Rudnicky - The coordinator of Hephaestus. Gave a lot of valuable 113suggestions to Sphinx and various open source speech projects of CMU. 114 115Jack Mostow - The principal investigator of project LISTEN. A lot of 116features of Sphinx are motivated by project LISTEN. From example, 117FSM's version of Sphinx 2. 118 119Jim Baker - Interaction with him benefits a lot of young researchers 120and develoeprs in CMU's speech group. 121 122Raj Reddy - For his vision in speech recognition and his good sense of 123choosing the name "Sphinx". 124 125We would also like to thank to all users who have kindly feedbacked to 126us and make Sphinx's better. 127 128To make this list complete, we may need to put all the names of CMU's 129graduates and faculties here. 130 131 132Others we'd like to thank 133 134Dr Takuji Nishimura and Dr Makoto Matsumoto. - Thanks to their effort 135in open sourcing a Mersene Twister prime-based random generator, we 136now have an extremely good portable random generator. 137 138