Picture for Huahuan Zheng

Huahuan Zheng

An Empirical Study of Language Model Integration for Transducer based Speech Recognition

Add code
Mar 31, 2022
Figure 1 for An Empirical Study of Language Model Integration for Transducer based Speech Recognition
Figure 2 for An Empirical Study of Language Model Integration for Transducer based Speech Recognition
Figure 3 for An Empirical Study of Language Model Integration for Transducer based Speech Recognition
Viaarxiv icon

CUSIDE: Chunking, Simulating Future Context and Decoding for Streaming ASR

Add code
Mar 31, 2022
Figure 1 for CUSIDE: Chunking, Simulating Future Context and Decoding for Streaming ASR
Figure 2 for CUSIDE: Chunking, Simulating Future Context and Decoding for Streaming ASR
Figure 3 for CUSIDE: Chunking, Simulating Future Context and Decoding for Streaming ASR
Figure 4 for CUSIDE: Chunking, Simulating Future Context and Decoding for Streaming ASR
Viaarxiv icon

Multilingual and crosslingual speech recognition using phonological-vector based phone embeddings

Add code
Jul 11, 2021
Figure 1 for Multilingual and crosslingual speech recognition using phonological-vector based phone embeddings
Figure 2 for Multilingual and crosslingual speech recognition using phonological-vector based phone embeddings
Figure 3 for Multilingual and crosslingual speech recognition using phonological-vector based phone embeddings
Figure 4 for Multilingual and crosslingual speech recognition using phonological-vector based phone embeddings
Viaarxiv icon

Advancing CTC-CRF Based End-to-End Speech Recognition with Wordpieces and Conformers

Add code
Jul 08, 2021
Figure 1 for Advancing CTC-CRF Based End-to-End Speech Recognition with Wordpieces and Conformers
Figure 2 for Advancing CTC-CRF Based End-to-End Speech Recognition with Wordpieces and Conformers
Figure 3 for Advancing CTC-CRF Based End-to-End Speech Recognition with Wordpieces and Conformers
Figure 4 for Advancing CTC-CRF Based End-to-End Speech Recognition with Wordpieces and Conformers
Viaarxiv icon

Efficient Neural Architecture Search for End-to-end Speech Recognition via Straight-Through Gradients

Add code
Nov 11, 2020
Figure 1 for Efficient Neural Architecture Search for End-to-end Speech Recognition via Straight-Through Gradients
Figure 2 for Efficient Neural Architecture Search for End-to-end Speech Recognition via Straight-Through Gradients
Figure 3 for Efficient Neural Architecture Search for End-to-end Speech Recognition via Straight-Through Gradients
Figure 4 for Efficient Neural Architecture Search for End-to-end Speech Recognition via Straight-Through Gradients
Viaarxiv icon

An empirical study of domain-agnostic semi-supervised learning via energy-based models: joint-training and pre-training

Add code
Oct 25, 2020
Figure 1 for An empirical study of domain-agnostic semi-supervised learning via energy-based models: joint-training and pre-training
Figure 2 for An empirical study of domain-agnostic semi-supervised learning via energy-based models: joint-training and pre-training
Figure 3 for An empirical study of domain-agnostic semi-supervised learning via energy-based models: joint-training and pre-training
Figure 4 for An empirical study of domain-agnostic semi-supervised learning via energy-based models: joint-training and pre-training
Viaarxiv icon