REVIEW 2 cited by
Pruned RNN-T for fast, memory-efficient ASR training
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
The RNN-Transducer (RNN-T) framework for speech recognition has been growing in popularity, particularly for deployed real-time ASR systems, because it combines high accuracy with naturally streaming recognition. One of the drawbacks of RNN-T is that its loss function is relatively slow to compute, and can use a lot of memory. Excessive GPU memory usage can make it impractical to use RNN-T loss in cases where the vocabulary size is large: for example, for Chinese character-based ASR. We introduce a method for faster and more memory-efficient RNN-T loss computation. We first obtain pruning bounds for the RNN-T recursion using a simple joiner network that is linear in the encoder and decoder embeddings; we can evaluate this without using much memory. We then use those pruning bounds to evaluate the full, non-linear joiner network.
Forward citations
Cited by 2 Pith papers
-
Early Attentive Sparsification Accelerates Neural Speech Transcription
Attention-based early audio-token sparsification at 40-60% sparsity accelerates Whisper ASR up to 1.6x with under 1% relative WER loss, across ten model variants, with no fine-tuning.
-
TokenVerse++: Towards Flexible Multitask Learning with Dynamic Task Activation
Adding task-specific learned vectors to acoustic embeddings lets a transducer ASR model train on partially labeled data, matching or beating the fully labeled TokenVerse baseline on most tasks.
Discussion (0). Continue with ORCID to comment.