About 22 results
Open links in new tab
  1. nttcslab/time-considerable-dialogue-model - GitHub

    This repository provides the evaluation datasets used in our paper "Time-Considerable Dialogue Models via Reranking by Time Dependency" (EMNLP Findings 2023). In the datasets, tweets are utilized as …

  2. GitHub - nttcslab/m2d: Masked Modeling Duo: Towards a Universal …

    Masked Modeling Duo: Towards a Universal Audio Pre-training Framework - nttcslab/m2d

  3. Activity · nttcslab/cone · GitHub

    Official PyTorch Implementation of "Deep Quantigraphic Image Enhancement via Comparametric Equations" (ICASSP2023) - Activity · nttcslab/cone

  4. GitHub - nttcslab/adaptive-leveling-BO

    title = {Stoichiometric growth of SrTiO$_3$ films via {Bayesian} optimization with adaptive prior mean}, author = {Yuki K. Wakabayashi and Takuma Otsuka and Yoshiharu Krockenberger and Hiroshi …

  5. EVAR ~ Evaluation package for Audio Representations - GitHub

    ft_lr: The learning rate (scheduled via cosine annealing) for fine-tuning (e.g., 0.001) ft_early_stop_epochs: The number of early stopping epochs, set to -1 to disable early stopping

  6. cone/exp/mit/train-20221108-052232/cem.txt at master - GitHub

    Nov 8, 2022 · Official PyTorch Implementation of "Deep Quantigraphic Image Enhancement via Comparametric Equations" (ICASSP2023) - cone/exp/mit/train-20221108-052232/cem.txt at master · …

  7. GitHub - nttcslab/dcase2023_task2_baseline_ae

    Contribute to nttcslab/dcase2023_task2_baseline_ae development by creating an account on GitHub.

  8. GitHub - nttcslab/composing-general-audio-repr: Composing General …

    The pre-trained weight will be loaded in the class AR_VGGish (evar/evar/ar_vggish.py), via the fusion wrapper class AR_VGGish_Fusion (to_evar/evar/ar_vggish_ext.py).

  9. GitHub

    Cnn14 fusion model","","CNN14-Fusion is implemented as a class GeneralPurposeCnn14 in [gp_cnn14.py] (gp_cnn14.py).","It has an extra parameter, `layers`, to specify which network blocks …

  10. GitHub - nttcslab/collision-probability-matching

    Note that this code only performs training under the proposed CPM loss (in addition to the cross-entropy loss) and excludes the calculation of the true collision probability, i.e., the test-retest reliability, and …