Efficient Speech Quality Assessment using Self-supervised Framewise Embeddings
This work addresses the need for faster iteration and deployment on resource-limited hardware for speech quality practitioners, contributing to sustainable machine learning.
The paper tackled the problem of automatic speech quality assessment by proposing an efficient system that achieves results comparable to the state-of-the-art model in the ConferencingSpeech 2022 challenge, with significant reductions in parameters (40-60x), FLOPS (100x), memory consumption (10-15x), and latency (30x).
Automatic speech quality assessment is essential for audio researchers, developers, speech and language pathologists, and system quality engineers. The current state-of-the-art systems are based on framewise speech features (hand-engineered or learnable) combined with time dependency modeling. This paper proposes an efficient system with results comparable to the best performing model in the ConferencingSpeech 2022 challenge. Our proposed system is characterized by a smaller number of parameters (40-60x), fewer FLOPS (100x), lower memory consumption (10-15x), and lower latency (30x). Speech quality practitioners can therefore iterate much faster, deploy the system on resource-limited hardware, and, overall, the proposed system contributes to sustainable machine learning. The paper also concludes that framewise embeddings outperform utterance-level embeddings and that multi-task training with acoustic conditions modeling does not degrade speech quality prediction while providing better interpretation.