Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 29 for “"TDNN"”.

  1. Nízko-dimenzionální faktorizace pro "End-To-End" řečové systémy

    … oneskorených neurónových sietí s faktorizáciou (TDNN-F) a bez nej (TDNN) v jazyku Pytorch. Následne je porovnávaná s už existujúcou implementáciou v nástroji Kaldi, kde boli dosiahnuté podobné výsledky v rámci experimentovania s rôznymi architektúrami. V poslednej kapitole popisujeme dopad …

    brno-tech Repository record for Nízko-dimenzionální faktorizace pro "End-To-End" řečové systémy (opens in a new tab)

  2. Comparison of Evolved Memory Mechanisms in Neural Networks

    … (1) RNN : Recurrent Neural Networks, (2) TDNN : Time-Delayed Neural Networks (feed-forward), and (3): FFWD: an external marker dropper/detector network (feed-forward). In this thesis, we ask which of the three mechanisms is the easiest to evolve? To answer this question, we set up a simple …

    tamu Repository record for Comparison of Evolved Memory Mechanisms in Neural Networks (opens in a new tab)

  3. Low latency modeling of temporal contexts for speech recognition

    … networks, termed time-delay neural networks (TDNNs). These sub-sampled TDNNs reduce the computation complexity by ~5x, compared to TDNNs, during frame randomized pre-training. These models are shown to be effective in modeling long-span temporal contexts, however there is a performance gap …

    jhu Repository record for Low latency modeling of temporal contexts for speech recognition (opens in a new tab)

  4. Development of Oxidative Lime Pretreatment and Shock Treatment to Produce Highly Digestible Lignocellulose for Biofuel and Ruminant Feed Applications

    … OLP increased the total digestible nutrients (TDNN; g nutrients digested/100 g organic matter) of corn stover from 51.9 (untreated) to 72.6. Adding in pre-washed corn stover solubles to produce a combined feed (17.8 percent corn stover solubles and 82.2 percent shock OLP corn stover) increased …

    tamu Repository record for Development of Oxidative Lime Pretreatment and Shock Treatment to Produce Highly Digestible Lignocellulose for Biofuel and Ruminant Feed Applications (opens in a new tab)

  5. Systém pro automatické rozpoznávání českých nářečí

    … baseline modely pomocí ResNet a ECAPA-TDNN, než přistoupíme k doladění self-supervised modelu WavLM Base+ s multi-headed factorized attentive (MHFA) pooling backendem. Nejlepší LID systém dosahuje 94,9 % přesnosti na vývojové sadě VoxLingua107, čímž překonává baseline ECAPA-TDNN a XLSR. …

    brno-tech Repository record for Systém pro automatické rozpoznávání českých nářečí (opens in a new tab)

  6. Performance Enhancement of Time Delay and Convolutional Neural Networks Employing Sparse Representation in the Transform Domains

    … (ANN) such as time delay neural networks (TDNN) and convolutional neural networks (CNNs). We show how TDNNs can be enhanced using a sparsifying transform layer that significantly improves learning time and forecasting performance for time series. We mathematically prove that the improvement …

    ucf Repository record for Performance Enhancement of Time Delay and Convolutional Neural Networks Employing Sparse Representation in the Transform Domains (opens in a new tab)

  7. A neural network approach for generating derivative information from quantized position measurements

    … the objectives using Tap Delay Neural Networks (TDNN). We incorporate the proposed methodology into the control system of a flexible joint manipulator. Numerous simulation results show the success of our proposed scheme within robot control loops. Additionally, we compare the performance of a …

    concordia Download PDF for A neural network approach for generating derivative information from quantized position measurements (opens in a new tab)

  8. PARAMETRICKY EFEKTIVNÍ MODELY PRO AUTOMATICKÉ ROZPOZNÁVÁNÍ ŘEČI A JAZYKA

    … Pro úlohu LID je tradiční diskriminátor typu TDNN nahrazen faktorizovanou neuronovou sítí (TDNN-F), čímž se dosahuje 30–50% redukce parametrů při zachování plné diskriminační kapacity. Dále práce zkoumá použití "Normalizing Flows" metody k modelování komplexních distribucí. To efektivně …

    brno-tech Repository record for PARAMETRICKY EFEKTIVNÍ MODELY PRO AUTOMATICKÉ ROZPOZNÁVÁNÍ ŘEČI A JAZYKA (opens in a new tab)

  9. Fault Detection and Isolation of Jet Engines Using Neural Networks

    … model (DNM) and time delay neural network (TDNN) are utilized. The DNM is constructed by using dynamic neurons which utilize infinite impulse response (IIR) filters to generate dynamical behaviour between the input and output of the network. On the other hand, TDNN uses several delays …

    concordia Download PDF for Fault Detection and Isolation of Jet Engines Using Neural Networks (opens in a new tab)

  10. Analýza kombinace informací ve více-kanálové verifikaci mluvčích

    … jako vstup do neuronové sítě (architektura ECAPA-TDNN), která extrahuje „embeddingy“, vektorovou reprezentaci mluvčího. Vektory jsou dále porovnány pomocí kosinové podobnosti, jehož výsledkem jsou skóre, reálná čísla. Nejlepšího relativního zlepšení proti jedno-kanálovým nahrávkám dosahuje fúze na …

    brno-tech Repository record for Analýza kombinace informací ve více-kanálové verifikaci mluvčích (opens in a new tab)

  11. Triukšmui atsparus AM radijo kalbos atpažinimo modelis su maršrutizavimu sąlygotais LoRA adapteriais /

    … methodology is introduced, combining an ECAPA-TDNN classification model used as an audio routing component with four distinct Whisper-Small LoRA adapters, each conditioned on specific noise pseudo-labels. This architecture allows even a relatively small model to achieve competitive …

    vilnius Repository record for Triukšmui atsparus AM radijo kalbos atpažinimo modelis su maršrutizavimu sąlygotais LoRA adapteriais / (opens in a new tab)

  12. Remaining Useful Life Prediction of Proton Exchange Membrane Fuel Cells Using Genetic Algorithm Based Nonlinear Autoregressive Exogenous Network

    … algorithm-based time delay neural network (GA-TDNN). The results show that the proposed method has better prediction accuracy compared with the other two artificial neural network algorithms.

    carleton Repository record for Remaining Useful Life Prediction of Proton Exchange Membrane Fuel Cells Using Genetic Algorithm Based Nonlinear Autoregressive Exogenous Network (opens in a new tab)

  13. Vocal style factorization for effective speaker recognition in affective scenarios

    … in the affective domain over baseline ECAPA-TDNN speaker recognition models. For instance, the true match rate at a false match rate of 1% improves from 27.6% to 46.2%. Additionally, we provide an analysis between speaker recognition match scores and emotions to identify challenging affective …

    msu Repository record for Vocal style factorization for effective speaker recognition in affective scenarios (opens in a new tab)

  14. From Voice to Vector: Design, Training and Applications of Speaker Embeddings

    … neurale netwerkarchitecturen, waaronder ECAPA-TDNN, die wereldwijd uitgroeiden tot een standaard in sprekerherkenning en meerdere internationale competities wonnen. Daarnaast werden trainingsmethoden bedacht om deze technologie robuuster en efficiënter te maken, en werd ze uitgebreid naar …

    ghent Repository record for From Voice to Vector: Design, Training and Applications of Speaker Embeddings (opens in a new tab)

  15. A Computer-Based Articulation Training Aid for Short Words (CATA)

    … An attempt at using a time-delay neural network (TDNN) classifier for distinguishing phonemes present in the CVCs is described. Experiments conducted to improve the VATA and the initial results obtained with the CVC display system are reported.

    odu Repository record for A Computer-Based Articulation Training Aid for Short Words (CATA) (opens in a new tab)

  16. Towards Automatic Analysis of Audio Recordings from Children with Autism Spectrum Disorder

    … that is applied to a time-delay neural network (TDNN) to improve the classification accuracy of recordings from infants and toddlers. One metric of interest for clinicians is the child’s response rate to questions from parents. We build an interrogative utterance detector that features a stack of …

    gatech Repository record for Towards Automatic Analysis of Audio Recordings from Children with Autism Spectrum Disorder (opens in a new tab)

  17. Child Speech Recognition as Low Resource Automatic Speech Recognition

    … factorized time delay neural networks (TDNN-F) are used as more data-efficient acoustic models (AM) for Deep Neural Network (DNN)-HMM hybrid ASR models; Adapting models trained on out-of-domain data: transfer learning is used to adapt end-to-end ASR model trained on adult speech for …

    jhu Repository record for Child Speech Recognition as Low Resource Automatic Speech Recognition (opens in a new tab)

  18. Computer analysis of children's non-native English speech for language learning and assessment

    … with conventional DNNs, LSTMs and more advanced TDNN models. This work uses the CALL-ST corpus and TLT-school corpus to study children's non-native English speech. Initially, data augmentation was explored on the CALL-ST corpus to address the lack of data problem using the AMI corpus and PF-STAR …

    birmingham Download PDF for Computer analysis of children's non-native English speech for language learning and assessment (opens in a new tab)

  19. Assessing the Impact of Codec-Induced Audio Degradation on Voice Biometric Systems

    … Speaker verification systems such as ECAPA-TDNN perform well on clean datasets, but their accuracy declines when low-bitrate codecs compress speech or when signals are resampled at reduced frequencies. In real-world deployments, systems adapt audio to bandwidth and storage limitations, often …

    rit Repository record for Assessing the Impact of Codec-Induced Audio Degradation on Voice Biometric Systems (opens in a new tab)

Page 1 of 2