Convolution-Based Attention Model with Positional Encoding for Streaming Speech Recognition on Embedded Devices
2021 IEEE Spoken Language Technology Workshop (SLT)（2021）
关键词
long filter-length,positional encoding,convolution-based ASR,encoder,short filter-length convolutional model,attention energy,end-to-end speech recognition task,convolution-based attention model,streaming speech recognition,embedded devices,server-based implementations,privacy protection,recurrent neural networks,long sequences,single-stream implementations,highly efficient convolutional model-based ASR,monotonic chunkwise attention,temporal convolution-based models,server-based ASR,on-device automatic speech recognition
AI 理解论文
溯源树
样例
生成溯源树，研究论文发展脉络
Chat Paper
正在生成论文摘要