Pooling classifier pre-trained using force-aligned phoneme and word labels on LibriSpeech

Reported on 3 benchmarks across 3 tasks · 1 paper · 3 SOTA

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

Speech2 results

DialogueonFluent Speech Commands
Accuracy (%)· uses extra data· 2019-04-07
98.8
best: 99.8 (Finstreder (Conformer + AMT, character-based))
SOTA
Speech Model Pre-training for End-to-End Spoken Language Understanding arXiv:1904.03670
Spoken Language UnderstandingonFluent Speech Commands
Accuracy (%)· uses extra data· 2019-04-07
98.8
best: 99.8 (Finstreder (Conformer + AMT, character-based))
SOTA
Speech Model Pre-training for End-to-End Spoken Language Understanding arXiv:1904.03670

Natural Language Processing1 result

Dialogue UnderstandingonFluent Speech Commands
Accuracy (%)· uses extra data· 2019-04-07
98.8
best: 99.8 (Finstreder (Conformer + AMT, character-based))
SOTA
Speech Model Pre-training for End-to-End Spoken Language Understanding arXiv:1904.03670