1 Introduction to the Edinburgh Speech Tools {#estintro}
2 ===========================================
6 The Edinburgh Speech Tools Library is a library of general speech software, written at the Centre
for Speech Technology Research at the University of Edinburgh.
8 The Edinburgh Speech Tools Library is written is C++ and provides a range of
9 tools
for common tasks found in speech processing. The library provides a
10 set of stand alone executable programs and a set of library calls
11 which can be linked into user programs.
13 Updates and news about the speech tools include releases and bug fixes
14 may be found at [Speech Tools website](http:
16 Most speech researchers spend a considerable amount of time writing,
17 developing and debugging code. In fact, many researchers spend most of
18 their time doing this. The sad fact is that most of
this time is spent
19 on unnecessary tasks - time which could be better spent doing
"real"
20 research. The library is intended to provide software that
21 programmers use day-to-day, and provide
this in an easy to use fashion.
23 # The Library {#thelibrary}
25 The Edinburgh Speech Tools Library has two main parts: a software
26 library and a set of programs which use the library.
28 A library is a single central place where useful software is kept. A
29 UNIX library is a single file (in
this case called `libestools.a`)
30 which can be linked to an individual program. When writing a program,
31 you can call any of the functions in the library, and they will
32 automatically be linked into your program when you compile. The key
33 point is that you never need look at the library itself or copy the
34 code in it. That way you can write small programs, concentrate on the
35 algorithms and not have to worry about any infrastructure issues.
37 The speech tools also provide a number of utility programs
for things
38 like playback, sampling rate conversion, file format conversion,
39 etc. Usually these programs are just wrap-around executables based on
40 standard speech tools library functions.
42 ## What does the library contain?
45 Includes tracks
for storing sets of time aligned coefficients, and waves
for
46 digitally sampled speech waveforms.
49 A comprenhensive system
for storing different kinds of linguistic information
50 is given. This is based on the Hetrogeneous Relation Graph
51 formalism ***CITE***. Feature structures, tress, lists, graphs
52 ect can all be represented with the linguistic classes.
55 Easy to use routines to record and play audio data without any fuss.
58 Commonly used signal processing algorithms such as including pitch tracking, cepstra and LPC, filtering, fourier analysis etc.
61 ### Statistical functions
66 Software support
for the Tilt intonation model
68 ### Speech Recognition
70 ### Utility Functions and Classes
71 Useful classes such as lists, vectors, matrices, strings and functions
for
72 reading files, parsing command lines etc.
74 ## Using the speechtools
75 Once installed, speech tools can be used either by running the
76 exectuable programs or by building your own C++ programs and linking
79 Documentation on executable programs is found in [Executable Programs](@ref estexec).
81 Instructions on how to build your own programs which use the library
82 are found in [Building speech tools](@ref estinstallbuild).