Picture for Dinghao Zhou

Dinghao Zhou

TouchTTS: An Embarrassingly Simple TTS Framework that Everyone Can Touch

Add code
Dec 12, 2024
Viaarxiv icon

U2++ MoE: Scaling 4.7x parameters with minimal impact on RTF

Add code
Apr 25, 2024
Figure 1 for U2++ MoE: Scaling 4.7x parameters with minimal impact on RTF
Figure 2 for U2++ MoE: Scaling 4.7x parameters with minimal impact on RTF
Figure 3 for U2++ MoE: Scaling 4.7x parameters with minimal impact on RTF
Figure 4 for U2++ MoE: Scaling 4.7x parameters with minimal impact on RTF
Viaarxiv icon

Say Goodbye to RNN-T Loss: A Novel CIF-based Transducer Architecture for Automatic Speech Recognition

Add code
Jul 27, 2023
Viaarxiv icon

Dynamic Latency for CTC-Based Streaming Automatic Speech Recognition With Emformer

Add code
Mar 29, 2022
Figure 1 for Dynamic Latency for CTC-Based Streaming Automatic Speech Recognition With Emformer
Figure 2 for Dynamic Latency for CTC-Based Streaming Automatic Speech Recognition With Emformer
Figure 3 for Dynamic Latency for CTC-Based Streaming Automatic Speech Recognition With Emformer
Figure 4 for Dynamic Latency for CTC-Based Streaming Automatic Speech Recognition With Emformer
Viaarxiv icon

Locality Matters: A Locality-Biased Linear Attention for Automatic Speech Recognition

Add code
Mar 29, 2022
Figure 1 for Locality Matters: A Locality-Biased Linear Attention for Automatic Speech Recognition
Figure 2 for Locality Matters: A Locality-Biased Linear Attention for Automatic Speech Recognition
Figure 3 for Locality Matters: A Locality-Biased Linear Attention for Automatic Speech Recognition
Figure 4 for Locality Matters: A Locality-Biased Linear Attention for Automatic Speech Recognition
Viaarxiv icon