The Meta-pi Network: Connectionist Rapid Adaptation for High-performance Multi-speaker Phoneme Recognition

We present a multi-network Time-Delay Neural Network (TDNN)based connectionist architecture that allows us to perform multispeaker phoneme discrimination (/b,d,gh at the speaker-dependent recognition rate of 98.4%. The overall network gates the phonemic decisions of modules trained on individual speakers to form its over-all classification decision. By dynamically adapting to the input speech and focusing on a combination of speaker-specific modules, the network outperforms a single TDNN trained on the speech of all six speakers (95.9%). To train this network we have developed a new form of multiplicative connection that we call the “Meta-Pi” connection. We illustrate how the Meta-Pi paradigm implements a dynamically adaptive Bayesian MAP classifier. It leams without supervision to recognize the speech of one particular speaker (99.8%) using a dynamic combination of intemal models of other speakers exclusively. The Meta-Pi model is a viable basis for a connectionist speech recognition system that can rapidly adapt to new speakers and vnrying speaker dialects.