Attention encoder-decoder model architecture is the backbone of several recent top performing foundation speech models: Whisper, Seamless, OWSM, and Canary-1B. However, the reported data and compute requirements for their training are prohibitive for many in the research community. In this work, we focus on the efficiency angle and ask the questions of whether we are training these speech models efficiently, and what can we do to improve? We argue that a major, if not the most severe, detrimental factor for training efficiency is related to the sampling strategy of sequential data. We show that negligence in mini-batch sampling leads to more than 50% computation being spent on padding. To that end, we study, profile, and optimize Canary-1B training to show gradual improvement in GPU utilization leading up to 5x increase in average batch sizes versus its original training settings. This in turn allows us to train an equivalent model using 4x less GPUs in the same wall time, or leverage the original resources and train it in 2x shorter wall time. Finally, we observe that the major inference bottleneck lies in the autoregressive decoder steps. We find that adjusting the model architecture to transfer model parameters from the decoder to the encoder results in a 3x inference speedup as measured by inverse real-time factor (RTFx) while preserving the accuracy and compute requirements for convergence. The training code and models will be available as open-source.
View on arXiv@article{żelasko2025_2503.05931,
title={ Training and Inference Efficiency of Encoder-Decoder Speech Models },
author={ Piotr Żelasko and Kunal Dhawan and Daniel Galvez and Krishna C. Puvvada and Ankita Pasad and Nithin Rao Koluguri and Ke Hu and Vitaly Lavrukhin and Jagadeesh Balam and Boris Ginsburg },
journal={arXiv preprint arXiv:2503.05931},
year={ 2025 }
}