LIVIVO - The Search Portal for Life Sciences

zur deutschen Oberfläche wechseln
Advanced search

Search results

Result 1 - 10 of total 11

Search options

  1. Book ; Online: Semi-supervised voice conversion with amortized variational inference

    Stephenson, Cory / Keskin, Gokce / Thomas, Anil / Elibol, Oguz H.

    2019  

    Abstract: In this work we introduce a semi-supervised approach to the voice conversion problem, in which speech from a source speaker is converted into speech of a target speaker. The proposed method makes use of both parallel and non-parallel utterances from the ... ...

    Abstract In this work we introduce a semi-supervised approach to the voice conversion problem, in which speech from a source speaker is converted into speech of a target speaker. The proposed method makes use of both parallel and non-parallel utterances from the source and target simultaneously during training. This approach can be used to extend existing parallel data voice conversion systems such that they can be trained with semi-supervision. We show that incorporating semi-supervision improves the voice conversion performance compared to fully supervised training when the number of parallel utterances is limited as in many practical applications. Additionally, we find that increasing the number non-parallel utterances used in training continues to improve performance when the amount of parallel training data is held constant.

    Comment: Accepted for publication at Interspeech 2019
    Keywords Statistics - Machine Learning ; Computer Science - Machine Learning ; Electrical Engineering and Systems Science - Audio and Speech Processing
    Subject code 006
    Publishing date 2019-09-30
    Publishing country us
    Document type Book ; Online
    Database BASE - Bielefeld Academic Search Engine (life sciences selection)

    More links

    Kategorien

  2. Book ; Online: Measuring the Effectiveness of Voice Conversion on Speaker Identification and Automatic Speech Recognition Systems

    Keskin, Gokce / Lee, Tyler / Stephenson, Cory / Elibol, Oguz H.

    2019  

    Abstract: This paper evaluates the effectiveness of a Cycle-GAN based voice converter (VC) on four speaker identification (SID) systems and an automated speech recognition (ASR) system for various purposes. Audio samples converted by the VC model are classified by ...

    Abstract This paper evaluates the effectiveness of a Cycle-GAN based voice converter (VC) on four speaker identification (SID) systems and an automated speech recognition (ASR) system for various purposes. Audio samples converted by the VC model are classified by the SID systems as the intended target at up to 46% top-1 accuracy among more than 250 speakers. This encouraging result in imitating the target styles led us to investigate if converted (synthetic) samples can be used to improve ASR training. Unfortunately, adding synthetic data to the ASR training set only marginally improves word and character error rates. Our results indicate that even though VC models can successfully mimic the style of target speakers as measured by SID systems, improving ASR training with synthetic data from VC systems needs further research to establish its efficacy.

    Comment: Accepted for publication at ICML 2019 Synthetic Realities: Workshop on Detecting Audio-Visual Fakes
    Keywords Electrical Engineering and Systems Science - Audio and Speech Processing
    Subject code 400
    Publishing date 2019-05-29
    Publishing country us
    Document type Book ; Online
    Database BASE - Bielefeld Academic Search Engine (life sciences selection)

    More links

    Kategorien

  3. Book ; Online: Shifted and Squeezed 8-bit Floating Point format for Low-Precision Training of Deep Neural Networks

    Cambier, Léopold / Bhiwandiwalla, Anahita / Gong, Ting / Nekuii, Mehran / Elibol, Oguz H / Tang, Hanlin

    2020  

    Abstract: Training with larger number of parameters while keeping fast iterations is an increasingly adopted strategy and trend for developing better performing Deep Neural Network (DNN) models. This necessitates increased memory footprint and computational ... ...

    Abstract Training with larger number of parameters while keeping fast iterations is an increasingly adopted strategy and trend for developing better performing Deep Neural Network (DNN) models. This necessitates increased memory footprint and computational requirements for training. Here we introduce a novel methodology for training deep neural networks using 8-bit floating point (FP8) numbers. Reduced bit precision allows for a larger effective memory and increased computational speed. We name this method Shifted and Squeezed FP8 (S2FP8). We show that, unlike previous 8-bit precision training methods, the proposed method works out-of-the-box for representative models: ResNet-50, Transformer and NCF. The method can maintain model accuracy without requiring fine-tuning loss scaling parameters or keeping certain layers in single precision. We introduce two learnable statistics of the DNN tensors - shifted and squeezed factors that are used to optimally adjust the range of the tensors in 8-bits, thus minimizing the loss in information due to quantization.
    Keywords Computer Science - Machine Learning
    Subject code 006
    Publishing date 2020-01-16
    Publishing country us
    Document type Book ; Online
    Database BASE - Bielefeld Academic Search Engine (life sciences selection)

    More links

    Kategorien

  4. Article ; Online: Scalable Nanogap Sensors for Non-Redox Enzyme Assays.

    Su, Xing / Tayebi, Noureddine / Credo, Grace M / Wu, Kai / Elibol, Oguz H / Liu, David J / Daniels, Jonathan S / Li, Handong / Hall, Drew A / Varma, Madoo

    ACS sensors

    2018  Volume 3, Issue 9, Page(s) 1773–1781

    Abstract: Clinical diagnostic assays that monitor redox enzyme activity are widely used in small, low-cost readout devices for point-of-care monitoring (e.g., a glucometer); however, monitoring non-redox enzymes in real-time using compact electronic devices ... ...

    Abstract Clinical diagnostic assays that monitor redox enzyme activity are widely used in small, low-cost readout devices for point-of-care monitoring (e.g., a glucometer); however, monitoring non-redox enzymes in real-time using compact electronic devices remains a challenge. We address this problem by using a highly scalable nanogap sensor array to observe electrochemical signals generated by a model non-redox enzyme system, the DNA polymerase-catalyzed incorporation of four modified, redox-tagged nucleotides. Using deoxynucleoside triphosphates (dNTPs) tagged with para-aminophenyl monophosphate (pAPP) to form pAP-deoxyribonucleoside tetra-phosphates (AP-dN4Ps), incorporation of the nucleotide analogs by DNA polymerase results in the release of redox inactive pAP-triphosphates (pAPP
    MeSH term(s) Alkaline Phosphatase/chemistry ; Aminophenols/chemistry ; Aniline Compounds/chemistry ; Animals ; Cattle ; DNA/chemistry ; DNA Polymerase I/analysis ; DNA Polymerase I/chemistry ; Electrochemical Techniques/methods ; Enzyme Assays/methods ; Escherichia coli/enzymology ; Escherichia coli Proteins/analysis ; Escherichia coli Proteins/chemistry ; Exodeoxyribonucleases/chemistry ; Nucleotides/chemistry ; Organophosphorus Compounds/chemistry ; Oxidation-Reduction
    Chemical Substances Aminophenols ; Aniline Compounds ; Escherichia coli Proteins ; Nucleotides ; Organophosphorus Compounds ; 4-aminophenylphosphate (72962-65-3) ; DNA (9007-49-2) ; DNA Polymerase I (EC 2.7.7.-) ; Exodeoxyribonucleases (EC 3.1.-) ; exodeoxyribonuclease III (EC 3.1.11.2) ; Alkaline Phosphatase (EC 3.1.3.1) ; 4-aminophenol (R7P8FRP05V)
    Language English
    Publishing date 2018-09-11
    Publishing country United States
    Document type Journal Article ; Research Support, Non-U.S. Gov't
    ISSN 2379-3694
    ISSN (online) 2379-3694
    DOI 10.1021/acssensors.8b00500
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

  5. Article ; Online: Surface immobilizable chelator for label-free electrical detection of pyrophosphate.

    Liu, David J / Credo, Grace M / Su, Xing / Wu, Kai / Lim, Hsiao C / Elibol, Oguz H / Bashir, Rashid / Varma, Madoo

    Chemical communications (Cambridge, England)

    2011  Volume 47, Issue 29, Page(s) 8310–8312

    Abstract: A new pyrophosphate (PPi) chelator was designed for surface-sensitive electrical detection of biomolecular reactions. This article describes the synthesis of the PPi-selective receptor, its surface immobilization and application to label-free electrical ... ...

    Abstract A new pyrophosphate (PPi) chelator was designed for surface-sensitive electrical detection of biomolecular reactions. This article describes the synthesis of the PPi-selective receptor, its surface immobilization and application to label-free electrical detection on a silicon-based field-effect transistor (FET) sensor.
    MeSH term(s) Catechols/chemistry ; Chelating Agents/chemistry ; Diphosphates/analysis ; Electrochemical Techniques/methods ; Fluorescent Dyes/chemistry ; Silicon/chemistry ; Surface Properties ; Transistors, Electronic
    Chemical Substances Catechols ; Chelating Agents ; Diphosphates ; Fluorescent Dyes ; catechol (LF3AJ089DQ) ; Silicon (Z4152N8IUI)
    Language English
    Publishing date 2011-06-20
    Publishing country England
    Document type Journal Article
    ZDB-ID 1472881-3
    ISSN 1364-548X ; 1359-7345 ; 0009-241X
    ISSN (online) 1364-548X
    ISSN 1359-7345 ; 0009-241X
    DOI 10.1039/c1cc12073e
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

  6. Article: Surface immobilizable chelator for label-free electrical detection of pyrophosphate

    Liu, David J / Credo, Grace M / Su, Xing / Wu, Kai / Lim, Hsiao C / Elibol, Oguz H / Bashir, Rashid / Varma, Madoo

    Chemical communications. 2011 July 11, v. 47, no. 29

    2011  

    Abstract: A new pyrophosphate (PPi) chelator was designed for surface-sensitive electrical detection of biomolecular reactions. This article describes the synthesis of the PPi-selective receptor, its surface immobilization and application to label-free electrical ... ...

    Abstract A new pyrophosphate (PPi) chelator was designed for surface-sensitive electrical detection of biomolecular reactions. This article describes the synthesis of the PPi-selective receptor, its surface immobilization and application to label-free electrical detection on a silicon-based field-effect transistor (FET) sensor.
    Keywords chelating agents ; chemical compounds ; chemical reactions ; transistors
    Language English
    Dates of publication 2011-0711
    Size p. 8310-8312.
    Publishing place The Royal Society of Chemistry
    Document type Article
    ZDB-ID 1472881-3
    ISSN 1364-548X ; 1359-7345 ; 0009-241X
    ISSN (online) 1364-548X
    ISSN 1359-7345 ; 0009-241X
    DOI 10.1039/c1cc12073e
    Database NAL-Catalogue (AGRICOLA)

    More links

    Kategorien

  7. Article ; Online: Label-free electrical detection of pyrophosphate generated from DNA polymerase reactions on field-effect devices.

    Credo, Grace M / Su, Xing / Wu, Kai / Elibol, Oguz H / Liu, David J / Reddy, Bobby / Tsai, Ta-Wei / Dorvel, Brian R / Daniels, Jonathan S / Bashir, Rashid / Varma, Madoo

    The Analyst

    2012  Volume 137, Issue 6, Page(s) 1351–1362

    Abstract: We introduce a label-free approach for sensing polymerase reactions on deoxyribonucleic acid (DNA) using a chelator-modified silicon-on-insulator field-effect transistor (SOI-FET) that exhibits selective and reversible electrical response to ... ...

    Abstract We introduce a label-free approach for sensing polymerase reactions on deoxyribonucleic acid (DNA) using a chelator-modified silicon-on-insulator field-effect transistor (SOI-FET) that exhibits selective and reversible electrical response to pyrophosphate anions. The chemical modification of the sensor surface was designed to include rolling-circle amplification (RCA) DNA colonies for locally enhanced pyrophosphate (PPi) signal generation and sensors with immobilized chelators for capture and surface-sensitive detection of diffusible reaction by-products. While detecting arrays of enzymatic base incorporation reactions is typically accomplished using optical fluorescence or chemiluminescence techniques, our results suggest that it is possible to develop scalable and portable PPi-specific sensors and platforms for broad biomedical applications such as DNA sequencing and microbe detection using surface-sensitive electrical readout techniques.
    MeSH term(s) Base Sequence ; Biosensing Techniques ; Chelating Agents/chemistry ; DNA/chemistry ; DNA/metabolism ; DNA-Directed DNA Polymerase/chemistry ; DNA-Directed DNA Polymerase/metabolism ; Diphosphates/analysis ; Electrochemical Techniques ; Molecular Sequence Data ; Molecular Structure ; Silanes/chemistry ; Silicon/chemistry ; Surface Properties ; Transistors, Electronic
    Chemical Substances Chelating Agents ; Diphosphates ; Silanes ; DNA (9007-49-2) ; DNA-Directed DNA Polymerase (EC 2.7.7.7) ; Silicon (Z4152N8IUI)
    Language English
    Publishing date 2012-01-19
    Publishing country England
    Document type Evaluation Study ; Journal Article ; Research Support, N.I.H., Extramural
    ZDB-ID 210747-8
    ISSN 1364-5528 ; 0003-2654
    ISSN (online) 1364-5528
    ISSN 0003-2654
    DOI 10.1039/c2an15930a
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

  8. Article: High-k dielectric Al₂O₃ nanowire and nanoplate field effect sensors for improved pH sensing

    Reddy, Bobby Jr / Dorvel, Brian R / Go, Jonghyun / Nair, Pradeep R / Elibol, Oguz H / Credo, Grace M / Daniels, Jonathan S / Chow, Edmond K. C / Su, Xing / Varma, Madoo / Alam, Muhammad A / Bashir, Rashid

    Biomedical microdevices. 2011 Apr., v. 13, no. 2

    2011  

    Abstract: Over the last decade, field-effect transistors (FETs) with nanoscale dimensions have emerged as possible label-free biological and chemical sensors capable of highly sensitive detection of various entities and processes. While significant progress has ... ...

    Abstract Over the last decade, field-effect transistors (FETs) with nanoscale dimensions have emerged as possible label-free biological and chemical sensors capable of highly sensitive detection of various entities and processes. While significant progress has been made towards improving their sensitivity, much is yet to be explored in the study of various critical parameters, such as the choice of a sensing dielectric, the choice of applied front and back gate biases, the design of the device dimensions, and many others. In this work, we present a process to fabricate nanowire and nanoplate FETs with Al₂O₃ gate dielectrics and we compare these devices with FETs with SiO₂ gate dielectrics. The use of a high-k dielectric such as Al₂O₃ allows for the physical thickness of the gate dielectric to be thicker without losing sensitivity to charge, which then reduces leakage currents and results in devices that are highly robust in fluid. This optimized process results in devices stable for up to 8 h in fluidic environments. Using pH sensing as a benchmark, we show the importance of optimizing the device bias, particularly the back gate bias which modulates the effective channel thickness. We also demonstrate that devices with Al₂O₃ gate dielectrics exhibit superior sensitivity to pH when compared to devices with SiO₂ gate dielectrics. Finally, we show that when the effective electrical silicon channel thickness is on the order of the Debye length, device response to pH is virtually independent of device width. These silicon FET sensors could become integral components of future silicon based Lab on Chip systems.
    Keywords nanowires ; width
    Language English
    Dates of publication 2011-04
    Size p. 335-344.
    Publisher Springer US
    Publishing place Boston
    Document type Article
    ZDB-ID 2004019-2
    ISSN 1572-8781 ; 1387-2176
    ISSN (online) 1572-8781
    ISSN 1387-2176
    DOI 10.1007/s10544-010-9497-z
    Database NAL-Catalogue (AGRICOLA)

    More links

    Kategorien

  9. Article ; Online: Silicon field effect transistors as dual-use sensor-heater hybrids.

    Reddy, Bobby / Elibol, Oguz H / Nair, Pradeep R / Dorvel, Brian R / Butler, Felice / Ahsan, Zahab / Bergstrom, Donald E / Alam, Muhammad A / Bashir, Rashid

    Analytical chemistry

    2011  Volume 83, Issue 3, Page(s) 888–895

    Abstract: We demonstrate the temperature mediated applications of a previously proposed novel localized dielectric heating method on the surface of dual purpose silicon field effect transistor (FET) sensor-heaters and perform modeling and characterization of the ... ...

    Abstract We demonstrate the temperature mediated applications of a previously proposed novel localized dielectric heating method on the surface of dual purpose silicon field effect transistor (FET) sensor-heaters and perform modeling and characterization of the underlying mechanisms. The FETs are first shown to operate as electrical sensors via sensitivity to changes in pH in ionic fluids. The same devices are then demonstrated as highly localized heaters via investigation of experimental heating profiles and comparison to simulation results. These results offer further insight into the heating mechanism and help determine the spatial resolution of the technique. Two important biosensor platform applications spanning different temperature ranges are then demonstrated: a localized heat-mediated DNA exchange reaction and a method for dense selective functionalization of probe molecules via the heat catalyzed complete desorption and reattachment of chemical functionalization to the transistor surfaces. Our results show that the use of silicon transistors can be extended beyond electrical switching and field-effect sensing to performing localized temperature controlled chemical reactions on the transistor itself.
    MeSH term(s) Catalysis ; Hot Temperature ; Microscopy, Electron, Scanning ; Silicon/chemistry ; Surface Properties ; Transistors, Electronic
    Chemical Substances Silicon (Z4152N8IUI)
    Language English
    Publishing date 2011-01-07
    Publishing country United States
    Document type Journal Article ; Research Support, N.I.H., Extramural ; Research Support, Non-U.S. Gov't ; Research Support, U.S. Gov't, Non-P.H.S.
    ZDB-ID 1508-8
    ISSN 1520-6882 ; 0003-2700
    ISSN (online) 1520-6882
    ISSN 0003-2700
    DOI 10.1021/ac102566f
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

  10. Article ; Online: High-k dielectric Al₂O₃ nanowire and nanoplate field effect sensors for improved pH sensing.

    Reddy, Bobby / Dorvel, Brian R / Go, Jonghyun / Nair, Pradeep R / Elibol, Oguz H / Credo, Grace M / Daniels, Jonathan S / Chow, Edmond K C / Su, Xing / Varma, Madoo / Alam, Muhammad A / Bashir, Rashid

    Biomedical microdevices

    2011  Volume 13, Issue 2, Page(s) 335–344

    Abstract: Over the last decade, field-effect transistors (FETs) with nanoscale dimensions have emerged as possible label-free biological and chemical sensors capable of highly sensitive detection of various entities and processes. While significant progress has ... ...

    Abstract Over the last decade, field-effect transistors (FETs) with nanoscale dimensions have emerged as possible label-free biological and chemical sensors capable of highly sensitive detection of various entities and processes. While significant progress has been made towards improving their sensitivity, much is yet to be explored in the study of various critical parameters, such as the choice of a sensing dielectric, the choice of applied front and back gate biases, the design of the device dimensions, and many others. In this work, we present a process to fabricate nanowire and nanoplate FETs with Al(2)O(3) gate dielectrics and we compare these devices with FETs with SiO(2) gate dielectrics. The use of a high-k dielectric such as Al(2)O(3) allows for the physical thickness of the gate dielectric to be thicker without losing sensitivity to charge, which then reduces leakage currents and results in devices that are highly robust in fluid. This optimized process results in devices stable for up to 8 h in fluidic environments. Using pH sensing as a benchmark, we show the importance of optimizing the device bias, particularly the back gate bias which modulates the effective channel thickness. We also demonstrate that devices with Al(2)O(3) gate dielectrics exhibit superior sensitivity to pH when compared to devices with SiO(2) gate dielectrics. Finally, we show that when the effective electrical silicon channel thickness is on the order of the Debye length, device response to pH is virtually independent of device width. These silicon FET sensors could become integral components of future silicon based Lab on Chip systems.
    MeSH term(s) Aluminum Oxide/chemistry ; Electric Impedance ; Hydrogen-Ion Concentration ; Nanowires/chemistry ; Reproducibility of Results ; Silicon/chemistry ; Transistors, Electronic
    Chemical Substances Aluminum Oxide (LMI26O6933) ; Silicon (Z4152N8IUI)
    Language English
    Publishing date 2011-01-04
    Publishing country United States
    Document type Journal Article ; Research Support, N.I.H., Extramural ; Research Support, Non-U.S. Gov't
    ZDB-ID 2004019-2
    ISSN 1572-8781 ; 1387-2176
    ISSN (online) 1572-8781
    ISSN 1387-2176
    DOI 10.1007/s10544-010-9497-z
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

To top