Picture for Rama Doddipatla

Rama Doddipatla

Self-regularised Minimum Latency Training for Streaming Transformer-based Speech Recognition

Apr 24, 2023
Mohan Li, Rama Doddipatla, Catalin Zorila

Figure 1 for Self-regularised Minimum Latency Training for Streaming Transformer-based Speech Recognition

Figure 2 for Self-regularised Minimum Latency Training for Streaming Transformer-based Speech Recognition

Figure 3 for Self-regularised Minimum Latency Training for Streaming Transformer-based Speech Recognition

Figure 4 for Self-regularised Minimum Latency Training for Streaming Transformer-based Speech Recognition

* 5 pages, 2 figures, accepted at Interspeech2022

Access Paper or Ask Questions

Non-autoregressive End-to-end Approaches for Joint Automatic Speech Recognition and Spoken Language Understanding

Apr 21, 2023
Mohan Li, Rama Doddipatla

Figure 1 for Non-autoregressive End-to-end Approaches for Joint Automatic Speech Recognition and Spoken Language Understanding

Figure 2 for Non-autoregressive End-to-end Approaches for Joint Automatic Speech Recognition and Spoken Language Understanding

Figure 3 for Non-autoregressive End-to-end Approaches for Joint Automatic Speech Recognition and Spoken Language Understanding

* 8 pages, 1 figure, accepted at IEEE SLT2023

Access Paper or Ask Questions

Multiple-hypothesis RNN-T Loss for Unsupervised Fine-tuning and Self-training of Neural Transducer

Jul 29, 2022
Cong-Thanh Do, Mohan Li, Rama Doddipatla

Figure 1 for Multiple-hypothesis RNN-T Loss for Unsupervised Fine-tuning and Self-training of Neural Transducer

Figure 2 for Multiple-hypothesis RNN-T Loss for Unsupervised Fine-tuning and Self-training of Neural Transducer

Figure 3 for Multiple-hypothesis RNN-T Loss for Unsupervised Fine-tuning and Self-training of Neural Transducer

Figure 4 for Multiple-hypothesis RNN-T Loss for Unsupervised Fine-tuning and Self-training of Neural Transducer

* Accepted to Interspeech 2022

Access Paper or Ask Questions

Speaker Reinforcement Using Target Source Extraction for Robust Automatic Speech Recognition

May 09, 2022
Catalin Zorila, Rama Doddipatla

Figure 1 for Speaker Reinforcement Using Target Source Extraction for Robust Automatic Speech Recognition

Figure 2 for Speaker Reinforcement Using Target Source Extraction for Robust Automatic Speech Recognition

Figure 3 for Speaker Reinforcement Using Target Source Extraction for Robust Automatic Speech Recognition

Figure 4 for Speaker Reinforcement Using Target Source Extraction for Robust Automatic Speech Recognition

* Accepted for ICASSP 2022

Access Paper or Ask Questions

On monoaural speech enhancement for automatic recognition of real noisy speech using mixture invariant training

May 03, 2022
Jisi Zhang, Catalin Zorila, Rama Doddipatla, Jon Barker

Figure 1 for On monoaural speech enhancement for automatic recognition of real noisy speech using mixture invariant training

Figure 2 for On monoaural speech enhancement for automatic recognition of real noisy speech using mixture invariant training

Figure 3 for On monoaural speech enhancement for automatic recognition of real noisy speech using mixture invariant training

Figure 4 for On monoaural speech enhancement for automatic recognition of real noisy speech using mixture invariant training

* Submitted to INTERSPEECH 2022

Access Paper or Ask Questions

Transformer-based Streaming ASR with Cumulative Attention

Mar 11, 2022
Mohan Li, Shucong Zhang, Catalin Zorila, Rama Doddipatla

Figure 1 for Transformer-based Streaming ASR with Cumulative Attention

Figure 2 for Transformer-based Streaming ASR with Cumulative Attention

Figure 3 for Transformer-based Streaming ASR with Cumulative Attention

Figure 4 for Transformer-based Streaming ASR with Cumulative Attention

* 5 pages, 1 figure, accepted at ICASSP 2022

Access Paper or Ask Questions

A study on cross-corpus speech emotion recognition and data augmentation

Jan 10, 2022
Norbert Braunschweiler, Rama Doddipatla, Simon Keizer, Svetlana Stoyanchev

Figure 1 for A study on cross-corpus speech emotion recognition and data augmentation

Figure 2 for A study on cross-corpus speech emotion recognition and data augmentation

Figure 3 for A study on cross-corpus speech emotion recognition and data augmentation

Figure 4 for A study on cross-corpus speech emotion recognition and data augmentation

* Accepted at ASRU 2021

Access Paper or Ask Questions

Monaural source separation: From anechoic to reverberant environments

Nov 15, 2021
Tobias Cord-Landwehr, Christoph Boeddeker, Thilo von Neumann, Catalin Zorila, Rama Doddipatla, Reinhold Haeb-Umbach

Figure 1 for Monaural source separation: From anechoic to reverberant environments

Figure 2 for Monaural source separation: From anechoic to reverberant environments

Figure 3 for Monaural source separation: From anechoic to reverberant environments

Figure 4 for Monaural source separation: From anechoic to reverberant environments

* Submitted for ICASSP 2022

Access Paper or Ask Questions

Towards Handling Unconstrained User Preferences in Dialogue

Sep 17, 2021
Suraj Pandey, Svetlana Stoyanchev, Rama Doddipatla

Figure 1 for Towards Handling Unconstrained User Preferences in Dialogue

Figure 2 for Towards Handling Unconstrained User Preferences in Dialogue

Figure 3 for Towards Handling Unconstrained User Preferences in Dialogue

Figure 4 for Towards Handling Unconstrained User Preferences in Dialogue

* 14 pages, 2 figures, The 12th International Workshop on Spoken Dialog System Technology

Access Paper or Ask Questions