Ibm Model-based and Frame-by-frame Speaker-recognition

Amidst the recent crazes for emerging technologies like speech recognition and biometrics, speaker recognition is slowly reaching the maturity to be deemed practical in many diierent applications. This paper presents new approaches for text-independent speaker recognition. The performances of the model-based algorithm presented concurrently at the ICASSP'98 conference and the frame-based algorithm presented in this paper are compared here. The engine, described here, provides multiple func-tionalities including those of identiication, veriication and classiication. The modes of operation and design choices allow for tight integration of the speech recognition and speaker recognition engines in a broad sense. This new architecture as well as the results obtained for very speciic tasks undoubtedly announce myriads of new applications where both technologies complement each other and can no longer be clearly distinguished as illustrated by the concept of speech biomet-rics. Hands-free and eyes-free human/machine transactions are moving a step further toward easier and more eecient interfaces and speech transactions are becoming more ubiquitous. nouvelle architecture de m^ eme que les r esultats obtenus pour des t^ aches bien sp eciiques annoncent une myri-ade de nouvelles applications o u ces deux technologies se compl ementent au point de ne plus ^ etre distingu es. Les interactions homme/machines en mode \mains li-bres et yeux libres" font un grand pas dans la directions d'interfaces plus ais es et plus eecaces. La parole de-vient davantage le vecteur pr eferr e de d'interaction.