Statements (34)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
deep learning model architecture
|
| gptkbp:architecture |
encoder
|
| gptkbp:author |
gptkb:Niki_Parmar
gptkb:Yonghui_Wu gptkb:Zhifeng_Chen Ruoming Pang Yu Zhang Anmol Gulati Chung-Cheng Chiu James Qin Jiahui Yu Shibo Wang Wei Han |
| gptkbp:combines |
convolutional neural networks
transformer networks |
| gptkbp:designedFor |
speech recognition
|
| https://www.w3.org/2000/01/rdf-schema#label |
Conformer
|
| gptkbp:improves |
gptkb:transformation
RNN-T automatic speech recognition accuracy |
| gptkbp:introduced |
gptkb:Google_Research
|
| gptkbp:introducedIn |
2020
|
| gptkbp:notablePublication |
Conformer: Convolution-augmented Transformer for Speech Recognition
|
| gptkbp:openSource |
gptkb:NVIDIA_NeMo
ESPnet TensorFlowASR |
| gptkbp:paperVenue |
Interspeech 2020
|
| gptkbp:usedIn |
ASR benchmarks
end-to-end speech recognition systems |
| gptkbp:uses |
self-attention mechanism
convolutional modules feed-forward modules |
| gptkbp:bfsParent |
gptkb:NeMo_ASR
|
| gptkbp:bfsLayer |
7
|