Picture for David Rybach

David Rybach

Large-scale Language Model Rescoring on Long-form Data

Jun 13, 2023
Figure 1 for Large-scale Language Model Rescoring on Long-form Data
Figure 2 for Large-scale Language Model Rescoring on Long-form Data
Figure 3 for Large-scale Language Model Rescoring on Long-form Data
Figure 4 for Large-scale Language Model Rescoring on Long-form Data
Viaarxiv icon

Alignment Entropy Regularization

Dec 22, 2022
Figure 1 for Alignment Entropy Regularization
Figure 2 for Alignment Entropy Regularization
Figure 3 for Alignment Entropy Regularization
Figure 4 for Alignment Entropy Regularization
Viaarxiv icon

E2E Segmentation in a Two-Pass Cascaded Encoder ASR Model

Nov 28, 2022
Figure 1 for E2E Segmentation in a Two-Pass Cascaded Encoder ASR Model
Figure 2 for E2E Segmentation in a Two-Pass Cascaded Encoder ASR Model
Figure 3 for E2E Segmentation in a Two-Pass Cascaded Encoder ASR Model
Figure 4 for E2E Segmentation in a Two-Pass Cascaded Encoder ASR Model
Viaarxiv icon

Global Normalization for Streaming Speech Recognition in a Modular Framework

May 26, 2022
Figure 1 for Global Normalization for Streaming Speech Recognition in a Modular Framework
Figure 2 for Global Normalization for Streaming Speech Recognition in a Modular Framework
Figure 3 for Global Normalization for Streaming Speech Recognition in a Modular Framework
Figure 4 for Global Normalization for Streaming Speech Recognition in a Modular Framework
Viaarxiv icon

E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR

Apr 22, 2022
Figure 1 for E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR
Figure 2 for E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR
Figure 3 for E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR
Figure 4 for E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR
Viaarxiv icon

Improving Rare Word Recognition with LM-aware MWER Training

Apr 15, 2022
Figure 1 for Improving Rare Word Recognition with LM-aware MWER Training
Figure 2 for Improving Rare Word Recognition with LM-aware MWER Training
Figure 3 for Improving Rare Word Recognition with LM-aware MWER Training
Figure 4 for Improving Rare Word Recognition with LM-aware MWER Training
Viaarxiv icon

Handling Compounding in Mobile Keyboard Input

Jan 17, 2022
Viaarxiv icon

Lookup-Table Recurrent Language Models for Long Tail Speech Recognition

Apr 09, 2021
Figure 1 for Lookup-Table Recurrent Language Models for Long Tail Speech Recognition
Figure 2 for Lookup-Table Recurrent Language Models for Long Tail Speech Recognition
Figure 3 for Lookup-Table Recurrent Language Models for Long Tail Speech Recognition
Figure 4 for Lookup-Table Recurrent Language Models for Long Tail Speech Recognition
Viaarxiv icon

Less Is More: Improved RNN-T Decoding Using Limited Label Context and Path Merging

Dec 12, 2020
Figure 1 for Less Is More: Improved RNN-T Decoding Using Limited Label Context and Path Merging
Figure 2 for Less Is More: Improved RNN-T Decoding Using Limited Label Context and Path Merging
Figure 3 for Less Is More: Improved RNN-T Decoding Using Limited Label Context and Path Merging
Figure 4 for Less Is More: Improved RNN-T Decoding Using Limited Label Context and Path Merging
Viaarxiv icon

A Streaming On-Device End-to-End Model Surpassing Server-Side Conventional Model Quality and Latency

Mar 28, 2020
Figure 1 for A Streaming On-Device End-to-End Model Surpassing Server-Side Conventional Model Quality and Latency
Figure 2 for A Streaming On-Device End-to-End Model Surpassing Server-Side Conventional Model Quality and Latency
Figure 3 for A Streaming On-Device End-to-End Model Surpassing Server-Side Conventional Model Quality and Latency
Figure 4 for A Streaming On-Device End-to-End Model Surpassing Server-Side Conventional Model Quality and Latency
Viaarxiv icon