On The Impact Of Quantization And Pruning Of Self-supervised Speech Models For Downstream Speech Recognition Tasks "in-the-wild''

Abstract

Recent advances with self-supervised learning have allowed speech recognition systems to achieve state-of-the-art (SOTA) word error rates (WER) while requiring only a fraction of the labeled training data needed by its predecessors. Notwithstanding, while such models achieve SOTA performance in matched train/test conditions, their performance degrades substantially when tested in unseen conditions. To overcome this problem, strategies such as data augmentation and/or domain shift training have been explored. Available models, however, are still too large to be considered for edge speech applications on resource-constrained devices, thus model compression tools are needed. In this paper, we explore the effects that train/test mismatch conditions have on speech recognition accuracy based on compressed self-supervised speech models. In particular, we report on the effects that parameter quantization and model pruning have on speech recognition accuracy based on the so-called robust wav2ve

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keypimentel2023on

Related papers

• Structured Pruning Of Self-supervised Pre-trained Models For Speech Recognition And Understanding (2023)11.39
• On The Transferability Of Whisper-based Representations For "in-the-wild" Cross-task Downstream Speech Applications (2023)0.00
• Distillation And Pruning For Scalable Self-supervised Representation-based Speech Quality Assessment (2025)8.09
• Speech Enhancement Using Self-supervised Pre-trained Model And Vector Quantization (2022)6.34
• Fine-tuning Strategies For Faster Inference Using Speech Self-supervised Models: A Comparative Study (2023)8.35
• A Noise-robust Self-supervised Pre-training Model Based Speech Representation Learning For Automatic Speech Recognition (2022)11.19
• Self-supervised Learning With Random-projection Quantizer For Speech Recognition (2022)0.00
• Usm-lite: Quantization And Sparsity Aware Fine-tuning For Speech Recognition With Universal Speech Models (2023)4.52