Picture for Deblin Bagchi

Deblin Bagchi

Low-resource Low-footprint Wake-word Detection using Knowledge Distillation

Add code
Jul 06, 2022
Figure 1 for Low-resource Low-footprint Wake-word Detection using Knowledge Distillation
Figure 2 for Low-resource Low-footprint Wake-word Detection using Knowledge Distillation
Figure 3 for Low-resource Low-footprint Wake-word Detection using Knowledge Distillation
Figure 4 for Low-resource Low-footprint Wake-word Detection using Knowledge Distillation
Viaarxiv icon

Perceptual Loss with Recognition Model for Single-Channel Enhancement and Robust ASR

Add code
Dec 11, 2021
Figure 1 for Perceptual Loss with Recognition Model for Single-Channel Enhancement and Robust ASR
Figure 2 for Perceptual Loss with Recognition Model for Single-Channel Enhancement and Robust ASR
Figure 3 for Perceptual Loss with Recognition Model for Single-Channel Enhancement and Robust ASR
Figure 4 for Perceptual Loss with Recognition Model for Single-Channel Enhancement and Robust ASR
Viaarxiv icon

Speech Synthesis as Augmentation for Low-Resource ASR

Add code
Dec 23, 2020
Figure 1 for Speech Synthesis as Augmentation for Low-Resource ASR
Figure 2 for Speech Synthesis as Augmentation for Low-Resource ASR
Viaarxiv icon

Phonetic Feedback for Speech Enhancement With and Without Parallel Speech Data

Add code
Mar 03, 2020
Figure 1 for Phonetic Feedback for Speech Enhancement With and Without Parallel Speech Data
Figure 2 for Phonetic Feedback for Speech Enhancement With and Without Parallel Speech Data
Figure 3 for Phonetic Feedback for Speech Enhancement With and Without Parallel Speech Data
Figure 4 for Phonetic Feedback for Speech Enhancement With and Without Parallel Speech Data
Viaarxiv icon

Spectral feature mapping with mimic loss for robust speech recognition

Add code
Mar 26, 2018
Figure 1 for Spectral feature mapping with mimic loss for robust speech recognition
Figure 2 for Spectral feature mapping with mimic loss for robust speech recognition
Figure 3 for Spectral feature mapping with mimic loss for robust speech recognition
Figure 4 for Spectral feature mapping with mimic loss for robust speech recognition
Viaarxiv icon