Design of a 2 × 4 Hybrid MMI-MZI Configuration with MMI Phase-Shifters

Speaker adaptation for hybrid MMI/connectionist speech-recognition systems

In this paper we present a new adaptation technique for our hybrid large vocabulary continuous speech recognition system. In most adaptation approaches the HMM parameters are reestimated. In our approach, however, we train a speaker independent continuous speech recognizer, then we keep the HMM parameters fixed and we train a second network, which transforms the features of the adaptation data ...

W3c Mmi Workshop Position Paper

In the user interface adaptation project, we focused on the device independent representation and adaptation of web application UIs. The project was conducted together with SAP Global Research & Innovation. Based on the findings of our research, we have successfully developed the Dialog Description Language (DDL, [1]) that allows device independent authoring of web content based on the single s...

Interaction The Audi MMI Interface

Approximations to the MMI criterion and their effect on lattice-based MMI

Although maximum mutual information (MMI) training has been used for hidden Markov model (HMM) parameter estimation for more than twenty years ([2], [8], [5], [9], and [14]), it has recently become an essential part of the acoustic modeling repertoire thanks to the refinements introduced by Woodland and Povey ([16] and [11]). The earliest incarnations of MMI worked well on small vocabulary task...

MMI filters configuration for dual-wavelength generation in a ring cavity erbium-doped fibre laser

Background: Dual wavelength laser generation has been of constant interest due to their applications in optical communications and teraheartz generation. A novel configuration for Dual-wavelength laser generation based on the use of a couple of multimode interference (MMI) filters is demonstrated in a ring cavity Erbium-doped fibre laser. Methods: The MMI filters consist of a segment of no-core...