Package: cmuclmtk Version: 0.99.0-0svn2 Architecture: armhf Maintainer: David Huggins-Daines Installed-Size: 436 Depends: perl, libc6 (>= 2.7), libcmuclmtk0 Homepage: http://cmusphinx.sourceforge.net/ Priority: extra Section: text Filename: pool/main/c/cmuclmtk/cmuclmtk_0.99.0-0svn2_armhf.deb Size: 125780 SHA256: ac65c88efecabe21694eb65e13362b1f3977dbd69a25f5067b19cb7bf6b692a6 SHA1: a7650f84c5e94109dd97f4aad501f3cc338c3a8e MD5sum: 86d3ec0ea176c2f651d78932f61a3206 Description: CMU-Cambridge Language Modeling Toolkit The CMU-Cambridge Language Modeling Toolkit is a free set of tools for constructing and testing statistical N-Gram language models. These models have various applications including speech recognition, machine translation, optical character and handwriting recognition. . This package contains the front-end tools for easy language model training as well as the basic tools for manipulating N-Gram and text files. Package: libcmuclmtk-dev Source: cmuclmtk Version: 0.99.0-0svn2 Architecture: armhf Maintainer: David Huggins-Daines Installed-Size: 326 Depends: libcmuclmtk0 (= 0.99.0-0svn2) Homepage: http://cmusphinx.sourceforge.net/ Priority: extra Section: libdevel Filename: pool/main/c/cmuclmtk/libcmuclmtk-dev_0.99.0-0svn2_armhf.deb Size: 53890 SHA256: 64e0d05e569f89ae3947e5683ed5236d2fd10ed74dac30fa5f4c214fec02e1e1 SHA1: b735fa32442053e2a1549238cb6e0b69cc621534 MD5sum: c984d4dcc48203442fd7f10a9982f807 Description: CMU-Cambridge Language Modeling Toolkit The CMU-Cambridge Language Modeling Toolkit is a free set of tools for constructing and testing statistical N-Gram language models. These models have various applications including speech recognition, machine translation, optical character and handwriting recognition. . This package contains the include files and libraries used to compile programs using the CMU-Cambridge Language Model Toolkit. Package: libcmuclmtk0 Source: cmuclmtk Version: 0.99.0-0svn2 Architecture: armhf Maintainer: David Huggins-Daines Installed-Size: 170 Depends: libc6 (>= 2.7) Homepage: http://cmusphinx.sourceforge.net/ Priority: extra Section: libs Filename: pool/main/c/cmuclmtk/libcmuclmtk0_0.99.0-0svn2_armhf.deb Size: 38346 SHA256: ea7dbeeffa0cb6758e2ff5e80157a6b2821e6f8f4a03e24b2b11c4b8def488f7 SHA1: ccd465af1dcf07775d9e3096f8a695d032c529c1 MD5sum: f5b18bd7fe31913597a8b3e3baaeb408 Description: CMU-Cambridge Language Modeling Toolkit The CMU-Cambridge Language Modeling Toolkit is a free set of tools for constructing and testing statistical N-Gram language models. These models have various applications including speech recognition, machine translation, optical character and handwriting recognition. . This package contains the shared library used by the CMU-Cambridge Language Model Toolkit. Package: libfst-dev Source: openfst Version: 1.3.4-1 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 50327 Depends: libfst1 (= 1.3.4-1) Homepage: http://www.openfst.org/ Priority: optional Section: libdevel Filename: pool/main/o/openfst/libfst-dev_1.3.4-1_armhf.deb Size: 3406860 SHA256: 894b34f26cdb8cf5319f4132b798cf9b364dbdeb3c54700fac39accb58ac2d22 SHA1: 1160ce05bb85f7441e4ea1c0109b5967e1a2da99 MD5sum: 70deda7810b630d498fd3f92946adb3d Description: weighted finite-state transducers library (development) OpenFst is a library for constructing, combining, optimizing, and searching weighted finite-state transducers (FSTs). Weighted finite-state transducers are automata where each transition has an input label, an output label, and a weight. The more familiar finite-state acceptor is represented as a transducer with each transition's input and output label equal. Finite-state acceptors are used to represent sets of strings (specifically, regular or rational sets); finite-state transducers are used to represent binary relations between pairs of strings (specifically, rational transductions). The weights can be used to represent the cost of taking a particular transition. . This package contains the development support files. Package: libfst-tools Source: openfst Version: 1.3.4-1 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 989 Depends: libfst1 (= 1.3.4-1), libc6 (>= 2.4), libgcc1 (>= 1:4.4.0), libstdc++6 (>= 4.4.0) Homepage: http://www.openfst.org/ Priority: optional Section: text Filename: pool/main/o/openfst/libfst-tools_1.3.4-1_armhf.deb Size: 183188 SHA256: 17ff950cbeabf3f81da2804cb64fd2981021d2fa0f2d48fba39f37443fc5987c SHA1: 38070ae832cb69411082ec54dc0a220cd00f9c27 MD5sum: 973232e46fd06863b2f2ec284d024a7b Description: weighted finite-state transducers library (tools) OpenFst is a library for constructing, combining, optimizing, and searching weighted finite-state transducers (FSTs). Weighted finite-state transducers are automata where each transition has an input label, an output label, and a weight. The more familiar finite-state acceptor is represented as a transducer with each transition's input and output label equal. Finite-state acceptors are used to represent sets of strings (specifically, regular or rational sets); finite-state transducers are used to represent binary relations between pairs of strings (specifically, rational transductions). The weights can be used to represent the cost of taking a particular transition. . This package provides the command line tools. Package: libfst1 Source: openfst Version: 1.3.4-1 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 11440 Depends: libc6 (>= 2.4), libgcc1 (>= 1:4.4.0), libstdc++6 (>= 4.9) Breaks: libfst-tools (<< 1.2) Homepage: http://www.openfst.org/ Priority: optional Section: libs Filename: pool/main/o/openfst/libfst1_1.3.4-1_armhf.deb Size: 1621024 SHA256: 9776d87f740d72e8771547077b75c655dd4684398a1497f6918663774e38567e SHA1: 57123af279ee73c956f3e635fba3b793dc2e28e7 MD5sum: 69bb0a466f09b98e6334d54b41e902e9 Description: weighted finite-state transducers library (runtime) OpenFst is a library for constructing, combining, optimizing, and searching weighted finite-state transducers (FSTs). Weighted finite-state transducers are automata where each transition has an input label, an output label, and a weight. The more familiar finite-state acceptor is represented as a transducer with each transition's input and output label equal. Finite-state acceptors are used to represent sets of strings (specifically, regular or rational sets); finite-state transducers are used to represent binary relations between pairs of strings (specifically, rational transductions). The weights can be used to represent the cost of taking a particular transition. . This package includes the shared library. Package: libfst1-dbg Source: openfst Version: 1.3.4-1 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 50222 Depends: libfst1 (= 1.3.4-1) Homepage: http://www.openfst.org/ Priority: extra Section: debug Filename: pool/main/o/openfst/libfst1-dbg_1.3.4-1_armhf.deb Size: 44843012 SHA256: 2324f375f85fc5e1fa3085bc2336e0c26f1ae83ad1ee5a04b5abfab80619774e SHA1: 174f34118c23564f4cc2f9442f9e5e7182c5a6fb MD5sum: 8794cb7c472ed6b6e382627308e8c9ef Description: weighted finite-state transducers library (debug symbols) OpenFst is a library for constructing, combining, optimizing, and searching weighted finite-state transducers (FSTs). . This package contains debug symbols. Package: libfst1-plugins-base Source: openfst Version: 1.3.4-1 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 3877 Depends: libfst1 (= 1.3.4-1), libc6 (>= 2.4), libgcc1 (>= 1:4.4.0), libstdc++6 (>= 4.6) Homepage: http://www.openfst.org/ Priority: optional Section: text Filename: pool/main/o/openfst/libfst1-plugins-base_1.3.4-1_armhf.deb Size: 377602 SHA256: c31afd60d2a4dcfc2fe49c4de84467b48fb094de494d6dc459acd1fb3d3c641b SHA1: 7666b5523516668730da35e470e89c5028676b77 MD5sum: 6e0c2e0cb415c76660bae5315ed0ae09 Description: weighted finite-state transducers library (base plugins) OpenFst is a library for constructing, combining, optimizing, and searching weighted finite-state transducers (FSTs). . This package contains base extensions that will be dynamically loaded into any binary compiled with OpenFst if the LD_LIBRARY_PATH (or equivalent) includes /usr/lib/fst. Package: libmitlm-dev Source: mitlm Version: 0.4.1-1 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 1635 Depends: libmitlm0 (= 0.4.1-1) Homepage: http://code.google.com/p/mitlm/ Priority: optional Section: libdevel Filename: pool/main/m/mitlm/libmitlm-dev_0.4.1-1_armhf.deb Size: 231404 SHA256: d175a2214606aeea0f2be520ab8cc44334a1f37b9cbba4c69f75850035490856 SHA1: 457d2cd443038eea2f4e164b47904a8c3fe25094 MD5sum: 3dd6900d1d7737c6f12d019948b07808 Description: MIT Language Modeling toolkit - development files The MIT Language Modeling (MITLM) toolkit is a set of tools designed for the efficient estimation of statistical n-gram language models involving iterative parameter estimation. It achieves much of its efficiency through the use of a compact vector representation of n-grams. . Currently, MITLM supports the following features: * Smoothing: Modified Kneser-Ney, Kneser-Ney, maximum likelihood * Interpolation: Linear interpolation, count merging, generalized linear interpolation * Evaluation: Perplexity * File formats: ARPA, binary, gzip, bz2 . This package provides development headers for MITLM. Package: libmitlm0 Source: mitlm Version: 0.4.1-1 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 558 Depends: libc6 (>= 2.7), libgcc1 (>= 1:4.4.0), libgfortran3 (>= 4.6), libstdc++6 (>= 4.4.0) Homepage: http://code.google.com/p/mitlm/ Priority: optional Section: libs Filename: pool/main/m/mitlm/libmitlm0_0.4.1-1_armhf.deb Size: 154850 SHA256: 8ba0c846907c231d318e4b4a53bc7d3340632f96d2009501214a2121336b3d35 SHA1: 941508690a73f1a5aa9486d680adaf3bc8914bf9 MD5sum: 5b0a407f7a8dd5367cd91348edb771da Description: MIT Language Modeling toolkit - runtime library The MIT Language Modeling (MITLM) toolkit is a set of tools designed for the efficient estimation of statistical n-gram language models involving iterative parameter estimation. It achieves much of its efficiency through the use of a compact vector representation of n-grams. . Currently, MITLM supports the following features: * Smoothing: Modified Kneser-Ney, Kneser-Ney, maximum likelihood * Interpolation: Linear interpolation, count merging, generalized linear interpolation * Evaluation: Perplexity * File formats: ARPA, binary, gzip, bz2 . This package contains the MITLM runtime library. Package: libmitlm0-dbg Source: mitlm Version: 0.4.1-1 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 1852 Depends: libmitlm0 (= 0.4.1-1) Homepage: http://code.google.com/p/mitlm/ Priority: extra Section: debug Filename: pool/main/m/mitlm/libmitlm0-dbg_0.4.1-1_armhf.deb Size: 1716322 SHA256: a8ce73c8d8f6081645e072dba44896607a86d653a14cd82b9cdfbfb0d26a4835 SHA1: 0b1a3c981d23e24c201e521d71eee75493a4e773 MD5sum: af4dfd3621d98d8296afa00a1efabc79 Description: MIT Language Modeling toolkit - debug symbols The MIT Language Modeling (MITLM) toolkit is a set of tools designed for the efficient estimation of statistical n-gram language models involving iterative parameter estimation. . This package provides the detached debug symbols for the MITLM library. Package: m2m-aligner Version: 1.2-1 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 308 Depends: libc6 (>= 2.15), libgcc1 (>= 1:4.4.0), libstdc++6 (>= 4.9) Built-Using: tclap (= 1.2.1-1) Homepage: http://code.google.com/p/m2m-aligner/ Priority: optional Section: science Filename: pool/main/m/m2m-aligner/m2m-aligner_1.2-1_armhf.deb Size: 108708 SHA256: e92706c416956647d95cebafd71a43f9b452be77e4367d881a9d5a336df5c8da SHA1: e111f7524a59244b08ff5e5fa6ef327a228bdce8 MD5sum: 5ece3b7bebcef17092c6e6efaec9c18a Description: many-to-many alignments for string transduction The algorithm implemented in m2m-aligner has been applied in letter-to-phoneme conversion, name transliteration and other tasks. In general, this algorithm creates lexicon alignments without requiring annotated data nor linguistic knowledge. .. Tarek Sherif originally proposed this algorithm as a part of his Mater thesis graduated in 2007 from University of Alberta. Sittichai Jiampojamarn later reimplemented this algorithm to the first version of m2m-aligner during its PhD at the department of Computing Science, University of Alberta. Since then, many refinements, improvements, and features have been included for various tasks. Package: mitlm Version: 0.4.1-1 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 228 Depends: libmitlm0 (= 0.4.1-1), libc6 (>= 2.7), libgcc1 (>= 1:4.4.0), libstdc++6 (>= 4.3.0) Homepage: http://code.google.com/p/mitlm/ Priority: optional Section: text Filename: pool/main/m/mitlm/mitlm_0.4.1-1_armhf.deb Size: 49804 SHA256: 52840a02ae065fbd6a9cf238218fcd0644f9dc82d5afc869d1075deb97e1914f SHA1: d0e9901cb9fcd59fd2011c342a392de4245ab492 MD5sum: bf819ca91095adf8dec72c71c26c9ba8 Description: MIT Language Modeling toolkit The MIT Language Modeling (MITLM) toolkit is a set of tools designed for the efficient estimation of statistical n-gram language models involving iterative parameter estimation. It achieves much of its efficiency through the use of a compact vector representation of n-grams. . Currently, MITLM supports the following features: * Smoothing: Modified Kneser-Ney, Kneser-Ney, maximum likelihood * Interpolation: Linear interpolation, count merging, generalized linear interpolation * Evaluation: Perplexity * File formats: ARPA, binary, gzip, bz2 . This package provides the command line tools. Package: phonetisaurus Version: 0.7.8-3 Architecture: armhf Maintainer: Giulio Paci Installed-Size: 3409 Depends: libc6 (>= 2.4), libfst1 (>= 1.3.3), libgcc1 (>= 1:4.4.0), libstdc++6 (>= 4.9), python:any Suggests: libfst-tools, mitlm, m2m-aligner, flite Built-Using: sparsehash (= 2.0.2-1) Homepage: http://code.google.com/p/phonetisaurus/ Priority: optional Section: science Filename: pool/main/p/phonetisaurus/phonetisaurus_0.7.8-3_armhf.deb Size: 549342 SHA256: 1a8904094aace1f0c2a1a7b8ea5e9ab05ea67fac2485772b1d49c017b24e3896 SHA1: 2004b47981e9769935ef11e655ed782c4a1c0814 MD5sum: 0a740ade2e4302c94163b728bfb1a08d Description: Grapheme to Phoneme conversion tool Phonetisaurus is a tool to perform Grapheme to Phoneme conversion that is it provides pronunciation (or spelling) hypotheses given a word's ortographic representation. It is possible to require more than one hypotheses for each word. Phonetisaurus is implemented as a Weighted Finite State Transducer.