Zobrazeno 1 - 10
of 64
pro vyhledávání: '"Huang, W. Ronny"'
Autor:
Huang, W. Ronny, Allauzen, Cyril, Chen, Tongzhou, Gupta, Kilol, Hu, Ke, Qin, James, Zhang, Yu, Wang, Yongqiang, Chang, Shuo-Yiin, Sainath, Tara N.
In the era of large models, the autoregressive nature of decoding often results in latency serving as a significant bottleneck. We propose a non-autoregressive LM-fused ASR system that effectively leverages the parallelization capabilities of acceler
Externí odkaz:
http://arxiv.org/abs/2401.12789
Autor:
Chen, Tongzhou, Allauzen, Cyril, Huang, Yinghui, Park, Daniel, Rybach, David, Huang, W. Ronny, Cabrera, Rodrigo, Audhkhasi, Kartik, Ramabhadran, Bhuvana, Moreno, Pedro J., Riley, Michael
Publikováno v:
ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
In this work, we study the impact of Large-scale Language Models (LLM) on Automated Speech Recognition (ASR) of YouTube videos, which we use as a source for long-form ASR. We demonstrate up to 8\% relative reduction in Word Error Eate (WER) on US Eng
Externí odkaz:
http://arxiv.org/abs/2306.08133
We propose a method of segmenting long-form speech by separating semantically complete sentences within the utterance. This prevents the ASR decoder from needlessly processing faraway context while also preventing it from missing relevant context wit
Externí odkaz:
http://arxiv.org/abs/2305.18419
Autor:
Huang, W. Ronny, Chang, Shuo-Yiin, Sainath, Tara N., He, Yanzhang, Rybach, David, David, Robert, Prabhavalkar, Rohit, Allauzen, Cyril, Peyser, Cal, Strohman, Trevor D.
We explore unifying a neural segmenter with two-pass cascaded encoder ASR into a single model. A key challenge is allowing the segmenter (which runs in real-time, synchronously with the decoder) to finalize the 2nd pass (which runs 900 ms behind real
Externí odkaz:
http://arxiv.org/abs/2211.15432
Autor:
Meng, Zhong, Chen, Tongzhou, Prabhavalkar, Rohit, Zhang, Yu, Wang, Gary, Audhkhasi, Kartik, Emond, Jesse, Strohman, Trevor, Ramabhadran, Bhuvana, Huang, W. Ronny, Variani, Ehsan, Huang, Yinghui, Moreno, Pedro J.
Publikováno v:
2022 IEEE Spoken Language Technology Workshop (SLT), Doha, Qatar
Text-only adaptation of a transducer model remains challenging for end-to-end speech recognition since the transducer has no clearly separated acoustic model (AM), language model (LM) or blank model. In this work, we propose a modular hybrid autoregr
Externí odkaz:
http://arxiv.org/abs/2210.17049
Autor:
Huang, W. Ronny, Chang, Shuo-yiin, Rybach, David, Prabhavalkar, Rohit, Sainath, Tara N., Allauzen, Cyril, Peyser, Cal, Lu, Zhiyun
Improving the performance of end-to-end ASR models on long utterances ranging from minutes to hours in length is an ongoing challenge in speech recognition. A common solution is to segment the audio in advance using a separate voice activity detector
Externí odkaz:
http://arxiv.org/abs/2204.10749
End-to-end (E2E) models are often being accompanied by language models (LMs) via shallow fusion for boosting their overall quality as well as recognition of rare words. At the same time, several prior works show that LMs are susceptible to unintentio
Externí odkaz:
http://arxiv.org/abs/2204.09606
Autor:
Huang, W. Ronny, Peyser, Cal, Sainath, Tara N., Pang, Ruoming, Strohman, Trevor, Kumar, Shankar
Language model fusion helps smart assistants recognize words which are rare in acoustic data but abundant in text-only corpora (typed search logs). However, such corpora have properties that hinder downstream performance, including being (1) too larg
Externí odkaz:
http://arxiv.org/abs/2203.05008
Capitalization normalization (truecasing) is the task of restoring the correct case (uppercase or lowercase) of noisy text. We propose a fast, accurate and compact two-level hierarchical word-and-character-based recurrent neural network model. We use
Externí odkaz:
http://arxiv.org/abs/2202.08171
Autor:
Li, Bo, Pang, Ruoming, Sainath, Tara N., Gulati, Anmol, Zhang, Yu, Qin, James, Haghani, Parisa, Huang, W. Ronny, Ma, Min, Bai, Junwen
Building ASR models across many languages is a challenging multi-task learning problem due to large variations and heavily unbalanced data. Existing work has shown positive transfer from high resource to low resource languages. However, degradations
Externí odkaz:
http://arxiv.org/abs/2104.14830