An Introductory Course on Speech Processing
Course notes and related files

Prof.
Thierry Dutoit

Course and conference Notes
Introduction au Traitement de la Parole, (pdf format), notes de cours pour le DEC2 en Ingénierie Linguistique, 60pp., T. Dutoit, FPMs, 2000.
"Je pense donc je suis?" - Un bilan des des développements récents en traitement automatique de la parole, (pdf format), notes de séminaires, 23pp., T. Dutoit, FPMs, 2000.
Traitement de la Parole R. BOITE, H. BOURLARD, T. DUTOIT, J. HANCQ, H. LEICH, 2000, 2nd Edition, 488 pp., Presses Polytechniques Universitaires Romandes, Lausanne, ISBN 2-88074-388-5.
Complements to the online course (pdf, 900 kB).

A (Short) Introduction to Speech Processing, T. DUTOIT, Tutorial for ICME2002, Lausanne, August 2002.
Introduction, TTS, ASR, Conclusion.

Course Transparencies
Introduction to Speech Processing, (pdf format), T. Dutoit, FPMs, 2000.
Speech Modeling, (pdf format), T. Dutoit, FPMs, 2000.
Speech Coding, (pdf format), T. Dutoit, FPMs, 2000.
Text-to-Speech Synthesis (short version).(pdf format), T. Dutoit, FPMs, 2000 (slides).
Text-to-Speech Synthesis, TTS part 2, TTS part 3, TTS part 4 (pdf format), T. Dutoit, FPMs, 2000.
Automatic Speech Recognition, (pdf format), T. Dutoit, FPMs, 2000.
Conclusion, (pdf format), T. Dutoit, FPMs, 2000.

Demo and Tutorial Programs Great list of efficient pointers on speech-related matters including free source codes for tutorials or projects.
Un vocodeur complet LPC sous Matlab, par M. Akkin, G. Lenoir, et B. Beuavais.
LPCLearn, Real-time interface to an LPC analysis/synthesis system - L.-M. Croisez, T. Dutoit, 2000. (You can also build on its Freeware Sources)
PetitChien, a toy, multispeaker, isolated word recognition system, (c) Interactive Speech Technologies, 2000.
The MBROLA project, which distributes free speech synthesizers in many languages, thanks to fruitful international collaborations.
The EULER project, which distributes free text-to-speech synthesizers.
MAD, une suite d'outils interactifs pour facilitéer la compréhension du traitement du signal (orienté signal acoustique), par Martin Cooke, Sheffied Univ.
VoiceBox, VoiceBox, a MATLAB toolbox for speech processing - Mike Brookes, 1998.
JAVA Applets illustrating elementary notions, such as f0, dB, harmonics, etc.

Audio files (analysis)
"Parenthèse", a French word sampled at 8 kHz, 16 bits, for tutorial analysis purposes.

Audio files (coding)
Example of LPC10 coding at 2400 bps.
Example of GSM-type MP-LPC coding at 13000 bps.
Example of CELP coding at 8000 bps.
Example of ADPCM coding at 16000 bps.

Audio files (synthesis)
Dudley's voder, 1939.
DEC(KLAT)Talk, 1983.
Infovox's rule-based synthesis, 1978.
Bell Labs's LPC-based TTS, 1980's.
France Telecom's PSOLA 1989.
Accuvoice's TTS system, 1997.
CSTR' Festival unit selection TTS system, 1997.
AT&T's non uniform unit selection TTS system, 1998.

Video files
Intro to speech synthesis, (c) Matiere Grise, RTBF, 1997.
Phonetics at ULB, (c) Matiere Grise, RTBF, 1997.
Mechanical synthesis from http://www.eng.kagawa-u.ac.jp/~sawada/.
R. Feynman's famous words on computer science, 1985.
The MBROLA video, (c) FPMs, 1997.
Testing the mobility of the stirrup, before surgery (Université de Tours).

Language Engineering, from A World of Understanding, Language Engineering showcase CD, Interactive Labs.
Life long learning, from A World of Understanding, Language Engineering showcase CD, Interactive Labs.
The information society , from A World of Understanding, Language Engineering showcase CD, Interactive Labs.
Language technologies , from A World of Understanding, Language Engineering showcase CD, Interactive Labs.

Pointers
Videos of TCTS/MULTITEL Lab, including the famous "I don't believe in Computer Science" by R. Feynman.
The unescapable Speech FAQ : all you have ever wanted to know about speech processing, but never dared asking.
Introduction au traitement automatique du langage (TAL), excellent site, avec de nombreux thèmes développés.
Promenade dans la cochlée (Univ. Montpellier).

Labs and Exercises
Travaux Pratiques de Traitement de la Parole, T. Dutoit, 2000.
The MATLAB primer, un document d'introduction à MATLAB, gratuit.

Student Projects (in French)
Voir cette page.

Reference Books

  • Techniques de Compression des Signaux, N. Moreau, Ed. Masson, 1995.
  • Le Traitement de la Parole, 2nd ed., R. Boite et al., presses polytechniques romandes, 2000.