A5064866374- Speech Recognition and Synthesis
- Speech and Audio Processing
- Monoclonal and Polyclonal Antibodies Research
- Glycosylation and Glycoproteins Research
- Amino Acid Enzymes and Metabolism
- Neuroscience and Neuropharmacology Research
- Speech and dialogue systems
- Ion channel regulation and function
- Polyamine Metabolism and Applications
- Immunotherapy and Immune Responses
- Epigenetics and DNA Methylation
- Advanced Data Compression Techniques
- Natural Language Processing Techniques
- RNA modifications and cancer
- Cancer, Hypoxia, and Metabolism
- HER2/EGFR in Cancer Research
- Music and Audio Processing
- Peptidase Inhibition and Analysis
- Cancer-related gene regulation
- Bladder and Urothelial Cancer Treatments
- Proteoglycans and glycosaminoglycans research
- Chemical Synthesis and Analysis
- Protein purification and stability
- Cancer Research and Treatments
- RNA Interference and Gene Delivery
Fujita Health University
2024-2025
Kindai University
2016-2025
Tohoku University
1994-2023
Tohoku Medical and Pharmaceutical University
1988-2023
Nihon University
2008-2019
Nihon Pharmaceutical University
2018-2019
Toshiba (Japan)
2006-2018
Osaka University of Pharmaceutical Sciences
2013
Keio University Hospital
2013
Kanazawa University
2013
This paper derives a speech parameter generation algorithm for HMM-based synthesis, in which the sequence is generated from HMMs whose observation vector consists of spectral and its dynamic feature vectors. In algorithm, we assume that state (state mixture multi-mixture case) or part unobservable (i.e., hidden latent). As result, iterates forward-backward case where given. Experimental results show by using can reproduce clear formant structure as compared with produced single-mixture HMMs.
This paper discusses a hidden Markov model (HMM) based on multi-space probability distribution (MSD). The HMMs are widely-used statistical models to characterize the sequence of speech spectra and have successfully been applied recognition systems. From these facts, it is considered that HMM useful for modeling pitch patterns speech. However, we cannot apply conventional discrete or continuous pattern since observation composed one-dimensional values symbol which represents "unvoiced"....
A statistical speech synthesis system based on the hidden Markov model (HMM) was recently proposed. In this system, spectrum, excitation, and duration of are modeled simultaneously by context-dependent HMMs, parameter vector sequences generated from HMMs themselves. This defines a problem in generative framework solves it maximum likelihood (ML) criterion. However, there is an inconsistency: although state probability density functions (PDFs) explicitly used part they have not been...
Abstract The targeting of antioxidant systems that allow stem-like cancer cells to avoid the adverse consequences oxidative stress might be expected improve efficacy treatment. Here, we show head and neck squamous cell carcinoma (HNSCC) express variant isoforms CD44 (CD44v) rely on activity cystine transporter subunit xCT for control their redox status. inhibition selectively induces apoptosis in CD44v-expressing tumor without affecting CD44v-negative differentiated same tumor. In contrast...
Similar to normal tissue stem cells, cancer cells (CSCs) are thought be quiescent or slow-cycling and, thereby, insensitive chemo- and radiotherapies. CD44, a cell surface component that interacts with the extracellular matrix, has been found highly expressed in CSCs of several solid tumors. However, relevancy between CD44(+) underlying mechanisms for emergence during tumorigenesis have not elucidated. Here we show gastric gland residing at squamo-columnar junction (SCJ) mouse stomach...
Extracellular free amino acids contribute to the interaction between a tumor and its microenvironment through effects on cellular metabolism malignant behavior. System xc(-) is composed of xCT CD98hc subunits functions as plasma membrane antiporter for uptake extracellular cystine in exchange intracellular glutamate. Here, we show that EGFR interacts with thereby promotes cell surface expression function human glioma cells. EGFR-expressing cells manifested both enhanced antioxidant capacity...
A large number of structurally diverse compounds act as open-channel blockers NMDA receptors. They may share discrete or overlapping binding sites within the channel. In this study, effects mutations in and around membrane-spanning pore-forming regions receptor subunits were studied with three blockers, MK-801, memantine, TB-3-4, using recombinant receptors expressed in<i>Xenopus laevis</i> oocytes. Mutations at critical asparagine residues M2 loop NR1 NR2B a tryptophan residue reduced block...
This paper describes improvements on the excitation model of an HMM-based text-to-speech system. In our previous work, natural sounding speech can be synthesized from trained HMMs. However, it has a typical quality “vocoded speech” since system uses traditional with either periodic impulse train or white noise. this paper, in order to reduce synthetic quality, mixed used MELP is incorporated into Excitation parameters are modeled by HMMs, and generated HMMs parameter generation algorithm...
There are complex interactions between spermine, protons, and ifenprodil at <i>N</i>-methyl-d-aspartate receptors. Spermine stimulation may involve relief of proton inhibition, whereas inhibition an increase in inhibition. We studied mutations acidic residues the NR1 subunit using voltage-clamp recording NR1/NR2B receptors expressed <i>Xenopus</i> oocytes. Mutations near site exon-5 insert, including E181 E185, reduced spermine Mutation NR1(D130N) sensitivity to by more than 500-fold, but...
In the present paper, a hidden-semi Markov model (HSMM) based speech synthesis system is proposed. hidden (HMM) which we have proposed, rhythm and tempo are controlled by state duration probability distributions modeled single Gaussian distributions. To speech, it constructs sentence HMM corresponding to an arbitralily given text determine durations maximizing their probabilities, then parameter vector sequence generated for sequence. However, there inconsistency: although synthesized from...
Describes a technique for synthesizing speech with arbitrary speaker characteristics using independent units, which we call "average voice" units. The is based on an HMM-based text-to-speech (TTS) system and maximum likelihood linear regression (MLLR) adaptation algorithm. In the TTS system, synthesis units are modeled by multi-space probability distribution (MSD) HMMs can model spectrum pitch simultaneously in unified framework. We derive extension of MLLR algorithm to apply it MSD-HMMs....
This paper presents a new text-to-speech synthesis system based on HMM which includes dynamic features, i.e., delta and delta-delta parameters of speech. The uses triphone HMMs as the units. share less than 2,000 clustered states, each is modelled by single Gaussian distribution. For given text to be synthesized, sentence constructed concatenating HMMs. Speech are generated from in such way that output probability maximized. speech signal synthesized directly obtained using mel log spectral...
This paper describes a technique for controlling the degree of expressivity desired emotional expression and/or speaking style synthesized speech in an HMM-based synthesis framework. With this technique, multiple expressions and styles are modeled single model by using multiple-regression hidden semi-Markov (MRHSMM). A set control parameters, called vector, is defined, each unit MRHSMM, which mean parameters state output duration distributions expressed vector. In stage, units modified...
Cancer development is often preceded by the appearance of preneoplastic lesions. In gastric carcinogenesis, chronic inflammation and histopathologic progression stomach epithelium lead to metaplasia eventually adenocarcinoma. The cell surface protein CD 44, especially its variant isoforms ( 44v), has been implicated in metaplasia–carcinoma sequence stomach. We recently found that 44v interacts with stabilizes x CT , a subunit cystine transporter system xc(–), cancer cells thereby increases...
Expression of CD 44, especially the variant isoforms ( 44v) this major cancer stem cell marker, contributes to reactive oxygen species ROS ) defense through stabilizing xCT (a cystine–glutamate transporter) and promoting glutathione synthesis. This enhances development increases chemotherapy resistance. We investigate role 44v in regulation system cholangiocarcinoma CCA ). Immunohistochemical staining p38 MAPK target) expression Opisthorchis viverrini ‐induced hamster tissues (at 60, 90,...
Abstract Targeting the function of membrane transporters in cancer stemlike cells is a potential new therapeutic approach. Cystine‐glutamate antiporter xCT expressed CD44 variant (CD44v)‐expressing contributes to resistance oxidative stress as well therapy through promoting glutathione (GSH)‐mediated antioxidant defense. Amino acid transport by might, thus, be promising target for treatment, whereas determination factors cell sensitivity xCT‐targeted remain unclear. Here, we demonstrate that...
<i>N</i>-Methyl-d-aspartate (NMDA) receptors are modulated by extracellular spermine and protons blocked in a voltage-dependent manner polyamine derivatives such as N<sup>1</sup>-dansyl-spermine (N<sup>1</sup>-DnsSpm). The effects of mutations the first third transmembrane domains (M1 M3) pore-forming loop (M2) NMDA receptor subunits were studied. Surprisingly, some M2 M3 NR1 subunit, including at W608 N616 M2, reduced stimulation proton inhibition. These may have long-range allosteric or...