Branchformer + GFSA

Reported on 4 benchmarks across 1 task · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio4 results

Speech RecognitiononLibriSpeech 100h test-other
Word Error Rate (WER)· 2023-12-07
22.25
SOTA
Graph Convolutions Enrich the Self-Attention in Transformers!arXiv:2312.04234
Speech RecognitiononLibriSpeech 100h test-clean
Word Error Rate (WER)· 2023-12-07
9.6
SOTA
Graph Convolutions Enrich the Self-Attention in Transformers!arXiv:2312.04234
Speech RecognitiononLibriSpeech test-clean
Word Error Rate (WER)· 2023-12-07
2.11
best: 0.985 (United Med ASR)
Graph Convolutions Enrich the Self-Attention in Transformers!arXiv:2312.04234
Speech RecognitiononLibriSpeech test-other
Word Error Rate (WER)· 2023-12-07
4.94
best: 2.48 (SAMBA ASR)
Graph Convolutions Enrich the Self-Attention in Transformers!arXiv:2312.04234