SERGIO CHARD PÉREZ
Area of knowledge
Theory of the Signal and Communications
Physics, Engineering of Systems and Theory of the Signal
Studies in which it gives
Technical engineering of Telecommunication. Sound and Image
This work publishes with a licence BY-NC-SA Creative Commons License
In this asignatura treat diverse appearances related with the digital treatment of the speech. They enter the following subjects: production of the voice, coding of the voice, recognition of the speech, synthesis of voice and recognition of the locutor.
|Archives||Educational resources||Resources web|
The general aim of the asignatura is to provide some basic knowledges on the problematic associated to the recognition and to the synthesis of the speech, as well as present the technicians and applications more employed. The asignatura supports in a lot of occasions in an exhaustive analysis of the signal of voice from the point of view of the theory of the signal, but does not centre only in these analyses but in the evaluation of the results to give him a more practical sense to the subject.
This generic aim centres in the following appearances:
- Understand the peculiarities of the signal of voice and of the physiological processes associated with his generation and catchment; understand how these peculiarities can influence in the methods of analysis.
- Enter the most common technicians of analysis and parametrisation of this signal and the physical sense that have.
- Understand our phonetics and relate the analytical concepts purchased with the study fonético of the speech.
- Present the technicians of coding of voice that has direct relation with his synthesis and with the own analyses of the already presented voice.
- Comprise all the problematic that affects to the production of voice of synthetic form and the technicians developed to face up them.
- Present the at present employed technicians in recognition of the speech and of the speaker and deepen in the difficulties of these types of analysis
- Present the practical applications of this field developed and to develop.
It treats of an asignatura of last course and optativa of an intensification; from this perspective have treated to give to the content a practical point of view: it does not treat so much to achieve that the student know to develop entirely a system of recognition or synthesis of the speech as that it know to evaluate which are the most employed systems in the actuality, which are his theoretical foundations and which his main advantages and inconvenient.
The temario of the asignatura can divide in three blocks:
Block 1: Generation and catchment of the signal of voice (subjects 1 and 2).
Block 2: Parametrisation, this is, as it can measure the signal of voice (subject 3).
Block 3: Systems of treatment of the signal of voice: (subjects 4, 5, 6 and 7)
- Coding: efficient forms to store/transmit the voice (ej. Of utility: mobile Telephony).
- Synthesis of the voice.: Allow that a machine speak.
- Recognition of voice: understand that it is saying .
- Recognition of the locutor: averiguar the identity of the person that speaks.
These are the fundamental points of each one of the subjects treated:
1. Production of voice and phonetics
1.1.- The signal of voice
1.2.- Characteristic temporary of the signal of voice
1.3.- Nature of the signals of voice
1.4.- Characteristic statistics of the signal of voice
1.5.- Models for analysis of the production of voice
1.6.- Acoustic phonetics
2. The perception of the voice.
2.1.- The sense of the audition
2.2.- Physiology of the auditory system
3. Analysis of the voice. Parametrisation.
3.1.- Analysis located of the voice
3.2.- Temporary analysis located
3.3.- Analysis located in frequency
3.4.- Analysis of linear prediction
3.5.- Spectral analysis located
4. Coding of the Voice.
4.1.- History of the codificadores of voice
4.2.- Sampling and quantification
4.3.- Classification of the codificadores of voice
4.4.- Codificadores Of form of wave
4.6.- Codificadores Hybrids
4.7.- Measured of quality
5. Synthesis of the Voice and of the Speech
5.2.- Conversion text-voice
5.3.- Processed linguistic
5.4.- Processed acoustic
5.5.- Evaluation of the quality
6. Recognition of voice
6.4.- Diagram of work
6.5.- Technical of recognition
7. Recognition of the locutor
7.3.- Phases of the recognition
7.4.- Technical of recognition
TEMARIO OF PRACTICES
During the development of the course there will be a group of practices to develop, which still are for determining. The subjects proposed for the practical are the following:
1. Capture of signal of voice; segmentation and classification
2. Parametrisation of the signal of voice captured
3. Reading of the spectrogram
4. Analysis of linear prediction
5. Síntetizador Of voice
6. Reconocedor Of voice
The realisation of the practices is conditioned to the evolution of the theory so that we will treat to adapt his distribution in the time so that it do not have to realise a practical whose theoretical foundations have not given still in the classes of theory.
|Summary: Subjects of Technicians of Recognition and Synthesis of the Speech.|
|RUA - Subject 1.ppt||Transparencies of the Subject 1||762Kb||Microsoft Powerpoint||Visualise/Open|
|RUA - Subject 2.ppt||Transparencies of the Subject 2||534Kb||Microsoft Powerpoint||Visualise/Open|
|RUA - Subject 3.ppt||Transparencies of the Subject 3||2542Kb||Microsoft Powerpoint||Visualise/Open|
|RUA - Subject 4.ppt||Transparencies of the Subject 4||1252Kb||Microsoft Powerpoint||Visualise/Open|
|RUA - Subject 5.ppt||Transparencies of the Subject 5||443Kb||Microsoft Powerpoint||Visualise/Open|
|RUA - Subject 6.ppt||Transparencies of the Subject 6||629Kb||Microsoft Powerpoint||Visualise/Open|
|RUA - Subject 7.ppt||Transparencies of the Subject 7||439Kb||Microsoft Powerpoint||Visualise/Open|
|RUA - Subject 1.Pdf||Transparencies of the Subject 1||755Kb||Season PDF||Visualise/Open|
|RUA - Subject 2.Pdf||Transparencies of the Subject 2||376Kb||Season PDF||Visualise/Open|
|RUA - Subject 3.Pdf||Transparencies of the Subject 3||1610Kb||Season PDF||Visualise/Open|
|RUA - Subject 4.Pdf||Transparencies of the Subject 4||334Kb||Season PDF||Visualise/Open|
|RUA - Subject 5.Pdf||Transparencies of the Subject 5||193Kb||Season PDF||Visualise/Open|
|RUA - Subject 6.Pdf||Transparencies of the Subject 6||280Kb||Season PDF||Visualise/Open|
|RUA - Subject 7.Pdf||Transparencies of the Subject 7||126Kb||Season PDF||Visualise/Open|
Furui, Sadaoki, Digital speech processing, syntesis and recognition, Marcel Dekker, New York , 2001.
Universidad de Alicante
Carretera de San Vicente del Raspeig s/n
03690 San Vicente del Raspeig
Tel: (+34) 96 590 3400Fax: (+34) 96 590 3464