Related papers: HuBERT-EE: Early Exiting HuBERT for Efficient Spee…

In this work, we analyzed and compared speech representations extracted from different frozen self-supervised learning (SSL) speech pre-trained models on their ability to capture articulatory features (AF) information and their subsequent…

The performance of voice-controlled systems is usually influenced by accented speech. To make these systems more robust, the frontend accent recognition (AR) technologies have received increased attention in recent years. As accent is a…

We show the performance of Automatic Speech Recognition (ASR) systems that use semi-supervised speech representations can be boosted by a complimentary pitch accent detection module, by introducing a joint ASR and pitch accent detection…

As NLP models become larger, executing a trained model requires significant computational resources incurring monetary and environmental costs. To better respect a given inference budget, we propose a modification to contextual…

Distributed DNN inference is becoming increasingly important as the demand for intelligent services at the network edge grows. By leveraging the power of distributed computing, edge devices can perform complicated and resource-hungry…

This paper introduces MauBERT, a multilingual extension of HuBERT that leverages articulatory features for robust cross-lingual phonetic representation learning. We continue HuBERT pre-training with supervision based on a…

We present a frontend for improving robustness of automatic speech recognition (ASR), that jointly implements three modules within a single model: acoustic echo cancellation, speech enhancement, and speech separation. This is achieved by…

Self-supervised pretraining on speech data has achieved a lot of progress. High-fidelity representation of the speech signal is learned from a lot of untranscribed data and shows promising performance. Recently, there are several works…

Self-supervised learning (SSL) speech models such as wav2vec and HuBERT have demonstrated state-of-the-art performance on automatic speech recognition (ASR) and proved to be extremely useful in low label-resource settings. However, the…

Recently, the end-to-end approach has been successfully applied to multi-speaker speech separation and recognition in both single-channel and multichannel conditions. However, severe performance degradation is still observed in the…

Early-Exit Large Language Models (EE-LLMs) enable high throughput inference by allowing tokens to exit early at intermediate layers. However, their throughput is limited by the computational and memory savings. Existing EE-LLM frameworks…

To achieve robust far-field automatic speech recognition (ASR), existing techniques typically employ an acoustic front end (AFE) cascaded with a neural transducer (NT) ASR model. The AFE output, however, could be unreliable, as the…

The attention-based end-to-end (E2E) automatic speech recognition (ASR) architecture allows for joint optimization of acoustic and language models within a single network. However, in a vanilla E2E ASR architecture, the decoder sub-network…

Whispering is an important mode of human speech, but no end-to-end recognition results for it were reported yet, probably due to the scarcity of available whispered speech data. In this paper, we present several approaches for end-to-end…

Acoustic word embeddings (AWEs) are vector representations of spoken words. An effective method for obtaining AWEs is the Correspondence Auto-Encoder (CAE). In the past, the CAE method has been associated with traditional MFCC features.…

Recent work has shown that it is possible to train a single model to perform joint acoustic echo cancellation (AEC), speech enhancement, and voice separation, thereby serving as a unified frontend for robust automatic speech recognition…

We present a novel approach to end-to-end automatic speech recognition (ASR) that utilizes pre-trained masked language models (LMs) to facilitate the extraction of linguistic information. The proposed models, BERT-CTC and BECTRA, are…

Effective spoken dialog systems should facilitate natural interactions with quick and rhythmic timing, mirroring human communication patterns. To reduce response times, previous efforts have focused on minimizing the latency in automatic…

Self-supervised learning (SSL) is a powerful tool that allows learning of underlying representations from unlabeled data. Transformer based models such as wav2vec 2.0 and HuBERT are leading the field in the speech domain. Generally these…

Machine learning models can solve complex tasks but often require significant computational resources during inference. This has led to the development of various post-training computation reduction methods that tackle this issue in…