Akbar M., Caelen J. (1998). Parole et traduction automatique : le module de reconnaissance RAPHAEL, COLLING-ACL'98 , Montreal (Quebec), 36-40. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
HALPIN : A Natural Language Information Retrieval System for a.. - Rouillard (1999) (Correct)
....We have shown that it is possible to dispatch through the Web a real time calculated speech synthesis, and we are now working on the integration of a voice recognition server module in our system. Some possibilities are currently being tested, such as the French Janus system [SCHULTZ 97] AKBAR 98] in order to allow the user a free dialogue with the machine, in a more natural and effective way. We also count on the insertion of a large thesaurus, such as the French Smiographe thesaurus from Memodata 6 , for a broader cover of the vocabulary in input as well as output. 6] ....
Akbar M., Caelen J. (1998). Parole et traduction automatique : le module de reconnaissance RAPHAEL, COLLING-ACL'98 , Montreal (Quebec), 36-40.
Internet Evolution And Progress In Full Automatic French.. - Vaufreydaz, Gery (2001) (2 citations) (Correct)
....evaluation material Now we will test different language models calculated on Web data in speech recognition task. We will conduct tests on 300 signals from the Aupelf evaluation of French dictation system [6] The vocabulary (used in 4.3) was given by the Aupelf. Signals contain OOV words. Raphal [7], our French recognition engine, is used in tests. Its acoustic models were trained on only 12 hours of speech. The computation of language models by the minimal block method is full automatic. The process takes in input a corpus, a vocabulary and the cut offs for bigrams and trigrams. In ....
Akbar M., Caelen J., Parole et traduction automatique : le module de reconnaissance RAPHAEL, COLLING-ACL'98, pp. 36-40, Montreal (Quebec), August 1998.
A New Methodology For Speech Corpora Definition.. - Vaufreydaz.. (2000) (2 citations) Self-citation (Akbar) (Correct)
....acoustic features. Because this acoustic modeling was not possible for us till now due to the lack of speech data to reliably train triphone models, these recorded corpora will allow us to build context dependent acoustic models for our spontaneous speech recognition module called RAPHAEL (Akbar Caelen, 1998). We are planning the recording session with more than one hundred speakers by the middle of April, 2000. 6. ....
Akbar M., Caelen J. (1998) "Parole et traduction automatique: le module de reconnaissance RAPHAEL", COLLING-ACL'98, pp. 36-40, Montreal (Quebec), August 1998.
A New Methodology For Speech Corpora Definition.. - Vaufreydaz.. (2000) (2 citations) Self-citation (Akbar) (Correct)
....acoustic features. Because this acoustic modeling was not possible for us till now due to the lack of speech data to reliably train triphone models, these recorded corpora will allow us to build context acoustic dependent models for our spontaneous speech recognition module called RAPHAEL (Akbar Caelen, 1998). We are planning the recording session with more than one hundred speakers by the middle of April, 2000. 6. ....
Akbar M., Caelen J. (1998) "Parole et traduction automatique: le module de reconnaissance RAPHAEL", COLLING-ACL'98, pp. 36-40, Montreal (Quebec), August 1998.
Internet Documents: A Rich Source for Spoken Language.. - Vaufreydaz, Akbar.. (1999) (1 citation) Self-citation (Akbar) (Correct)
....very rich source of information for language modeling. However to be useful, one has to clean and filter out the extracted documents based on application in which LMs will be used. We successfully applied this method by integrating a trained LM to our spontaneous speech recognition module (RAPHAEL [6]) We tested a modified version of this system in real situation during CSTAR [7] demonstrations. Furthermore, we work on a task oriented recognition engine adapted for information search and retrieval system, HALPIN [8] Integrating vocal technology offers ease of use and provides more natural ....
Akbar M., Caelen J., "Parole et traduction automatique: le module de reconnaissance RAPHAEL", COLLING-ACL'98, pp. 36-40, Montreal (Quebec), August 1998.
A Network Architecture for Building Applications That.. - Vaufreydaz.. (1999) (1 citation) Self-citation (Akbar) (Correct)
....On the other hand using a powerful server, the recognition latency is reduced to an acceptable delay for the users who are working on lightweight computers. Our recognition proxy is including in global architecture as shown in the Fig. 1. The recognition engine is based on RAPHAEL module [2] and currently supports French language. The connection manager is used to allow multiple connections to the same server or multiple servers at a time. As it is already explained it manages the queuing and routing of the recognition requests to the appropriate server and assures the recognition ....
AKBAR M., CAELEN J., Parole et traduction automatique: le module de reconnaissance RAPHAEL, COLLING-ACL'98, pp. 36-40, Montreal (Quebec), August 1998.
Halpin : A Multimodal And Conversational System For.. - Rouillard, Caelen Self-citation (Caelen) (Correct)
....interaction model. We have shown that is it possible to dispatch through the Web a real time calculated speech synthesis, and we are now working on the integration of a voice recognition server module in our system. Some possibilities are currently being tested, as the French Janus system [14] [1], in order to allow the user a free dialogue with the machine, in a more natural and effective way. We also count on the insertion of a large powerful thesaurus, as the French Dicologique thesaurus from [8] for a broader cover of the vocabulary in input as well as output. 4 ACKNOWLEDGEMENTS ....
AKBAR M., CAELEN J., Parole et traduction automatique : le module de reconnaissance RAPHAEL, COLLING-ACL'98, pp. 36-40, Montreal (Quebec), August 1998.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC