architecture · first seen 16 Oct 2026
recurrent neural network transducer
Also: RNN-T, transducer
A streaming speech recognition architecture combining an encoder, prediction network, and joint network trained with a transducer loss.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Flash RNN-T: Memory-Efficient Joint Network and Transducer Loss ComputationPapers · 16 Oct 2026 · Mahmoud Ashraf
- RoleHear: Pseudo-Labeling for Semi-Supervised Role-Attributed Streaming Speech RecognitionPapers · 24 Oct 2026 · Zefang Liu, Chenyang Zhu, Sangwoo Cho, Xujun Peng, Rongqing Huang, Shi-Xiong Zha…
- Fine-Tuning a Transducer for Impaired Speech: Where the Gains Appear and What They Mean for EvaluationPapers · 5 Nov 2026 · Kien Nguyen, Tuan Dinh, Bao Phan
- Gabor Gabor: E-Gaborformer, a Strictly Causal Speech Recognizer with Gabor-Dictionary Convolutions and Delta-Rule MemoryPapers · 5 Nov 2026 · Nir Krakowski, Tamir Cohen, Ohad Zohar, Chen Zarfati
Latest sources
- Gabor Gabor: E-Gaborformer, a Strictly Causal Speech Recognizer with Gabor-Dictionary Convolutions and Delta-Rule MemoryPapers · 5 Nov 2026 · Nir Krakowski, Tamir Cohen, Ohad Zohar, Chen Zarfati
- Fine-Tuning a Transducer for Impaired Speech: Where the Gains Appear and What They Mean for EvaluationPapers · 5 Nov 2026 · Kien Nguyen, Tuan Dinh, Bao Phan
- RoleHear: Pseudo-Labeling for Semi-Supervised Role-Attributed Streaming Speech RecognitionPapers · 24 Oct 2026 · Zefang Liu, Chenyang Zhu, Sangwoo Cho, Xujun Peng, Rongqing Huang, Shi-Xiong Zha…
- Flash RNN-T: Memory-Efficient Joint Network and Transducer Loss ComputationPapers · 16 Oct 2026 · Mahmoud Ashraf