We propose a new end-to-end neural acoustic model for automatic speech recognition. The model is composed of multiple blocks with residual connections between them. Each block consists of one or more modules with 1D time-channel separable convolutional layers, batch normalization, and ReLU layers. It is trained with CTC loss. The proposed network achieves near state-of-the-art accuracy on LibriSpeech and Wall Street Journal, while having fewer parameters than all competing models. We also demonstrate that this model can be effectively fine-tuned on new datasets.

No abstract

Constructing knowledge graphs from unstructured text is an important task that is relevant to many domains. Most previous work focuses on extracting information from sentences or paragraphs, due to the difficulty of analyzing longer contexts. In this paper we propose a new jointly trained model that can be used for various information extraction tasks at the document level. The tasks performed in this paper are entity and event identification, typing, and coreference resolution. In order to improve entity and event extraction, we utilize contextaware representations aggregated from the detected mentions of the corresponding entities and event triggers across the entire document. By extending our system to documentlevel, we can improve our results by incorporating cross-sentence dependencies and additional contextual information that might not be available at the sentence level, which allows for more globally optimized predictions. We evaluate our system on documents from the ACE05-E + dataset and find significant improvement over the sentence-level state-of-theart on entity extraction and event detection. 1

In this paper, we extend previous self-supervised approaches for language identification by experimenting with Conformer based architecture in a multilingual pre-training paradigm. We find that pre-trained speech models optimally encode language discriminatory information in lower layers. Further, we demonstrate that the embeddings obtained from these layers are significantly robust to classify unseen languages and different acoustic environments without additional training. After fine-tuning a pre-trained Conformer model on the VoxLin-gua107 dataset, we achieve results similar to current state-ofthe-art systems for language identification. More, our model accomplishes this with 5x less parameters. We open-source the model through the NVIDIA NeMo toolkit.

No abstract