Learn More
We apply the recently proposed Context-Dependent DeepNeural-Network HMMs, or CD-DNN-HMMs, to speech-to-text transcription. For single-pass speaker-independent recognition on the RT03S Fisher portion of phone-call transcription benchmark (Switchboard), the word-error rate is reduced from 27.4%, obtained by discriminatively trained Gaussian-mixture HMMs, to(More)
We investigate the potential of Context-Dependent Deep-Neural-Network HMMs, or CD-DNN-HMMs, from a feature-engineering perspective. Recently, we had shown that for speaker-independent transcription of phone calls (NIST RT03S Fisher data), CD-DNN-HMMs reduced the word error rate by as much as one third—from 27.4%, obtained by discriminatively trained(More)
Age Specific Human-Computer Interaction (ASHCI) has vast potential applications in daily life. However, automatic age estimation technique is still underdeveloped. One of the main reasons is that the aging effects on human faces present several unique characteristics which make age estimation a challenging task that requires non-standard classification(More)
The generation of induced pluripotent stem cells (iPSCs) and induced neuronal cells (iNCs) from somatic cells provides new avenues for basic research and potential transplantation therapies for neurological diseases. However, clinical applications must consider the risk of tumor formation by iPSCs and the inability of iNCs to self-renew in culture. Here we(More)
Apolipoprotein E4 (apoE4) is the major genetic risk factor for Alzheimer's disease. However, the underlying mechanisms are unclear. We found that female apoE4 knock-in (KI) mice had an age-dependent decrease in hilar GABAergic interneurons that correlated with the extent of learning and memory deficits, as determined in the Morris water maze, in aged mice.(More)
With the maturing of fourth generation (4G) standardization and the ongoing worldwide deployment of 4G cellular networks, research activities on 5G communication technologies have emerged in both the academic and industrial communities. Various organizations from different countries and regions have taken initiatives and launched programs aimed at potential(More)
We investigate back-propagation based sequence training of Context-Dependent Deep-Neural-Network HMMs, or CD-DNN-HMMs, for conversational speech transcription. Theoretically, sequence training integrates with backpropagation in a straight-forward manner. However, we find that to get reasonable results, heuristics are needed that point to a problem with(More)
We propose a novel regularized adaptation technique for context dependent deep neural network hidden Markov models (CD-DNN-HMMs). The CD-DNN-HMM has a large output layer and many large hidden layers, each with thousands of neurons. The huge number of parameters in the CD-DNN-HMM makes adaptation a challenging task, esp. when the adaptation set is small. The(More)
Recently, we developed context-dependent deep neural network (DNN) hidden Markov models for large vocabulary speech recognition. While reducing errors by 33% compared to its discriminatively trained Gaussian-mixture counterpart on the switchboard benchmark task, DNN requires much more parameters. In this paper, we report our recent work on DNN for improved(More)
Apolipoprotein (apo) E, a polymorphic protein with three isoforms (apoE2, apoE3, and apoE4), is essential for lipid homeostasis. Carriers of apoE4 are at higher risk for developing Alzheimer's disease. We have investigated adult neurogenesis in mice with knockout (KO) for apoE or with knockin (KI) alleles for human apoE3 or apoE4, and we report that(More)