Attention or Convolution: Transformer Encoders in Audio Language Models for Inference Efficiency

Sungho Jeon, Ching-Feng Yeh, Hakan Inan, Wei-Ning Hsu, Rashi Rungta, Yashar Mehdad, Daniel Bikel

Attention or Convolution: Transformer Encoders in Audio Language Models for Inference Efficiency: 5 upvotes on Hugging Face Daily Papers, #13 of 16 papers on 2023-11-07. Day-by-day upvote history.

In this paper, we show that a simple self-supervised pre-trained audio model can achieve comparable inference efficiency to more complicated pre-trained models with speech transformer encoders. These speech transformers rely on mixing convolutional modules with self-attention modules. They achieve state-of-the-art performance on ASR with top efficiency. We first show that employing these speech transformers as an encoder significantly improves the efficiency of pre-trained audio models as well. However, our study shows that we can achieve comparable efficiency with advanced self-attention solely. We demonstrate that this simpler approach is particularly beneficial with a low-bit weight quantization technique of a neural network to improve efficiency. We hypothesize that it prevents propagating the errors between different quantized modules compared to recent speech transformers mixing quantized convolution and the quantized self-attention modules.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.