paper-with-me

홈 › Papers

An Investigation of Monotonic Transducers for Large-Scale Automatic Speech Recognition

2022-04-19 · Niko Moritz, Frank Seide, Duc Le, Jay Mahadeokar, Christian Fuegen

The two most popular loss functions for streaming end-to-end automatic speech recognition (ASR) are RNN-Transducer (RNN-T) and connectionist temporal classification (CTC). Between these two loss types we can classify the monotonic RNN-T (MonoRNN-T) and the recently proposed CTC-like Transducer (CTC-T). Monotonic transducers have a few advantages. First, RNN-T can suffer from runaway hallucination, where a model keeps emitting non-blank symbols without advancing in time. Secondly, monotonic transducers consume exactly one model score per time step and are therefore more compatible with traditional FST-based ASR decoders. However, the MonoRNN-T so far has been found to have worse accuracy than RNN-T. It does not have to be that way: By regularizing the training via joint LAS training or parameter initialization from RNN-T, both MonoRNN-T and CTC-T perform as well or better than RNN-T. This is demonstrated for LibriSpeech and for a large-scale in-house data set.

📄 PDF Abstract BibTeX arXiv:2204.08858

Code (0)

등록된 구현이 없습니다.

Tasks

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Hallucinationspeech-recognitionSpeech Recognition

Similar Papers 제목 키워드 기반

Transducers with Pronunciation-aware Embeddings for Automatic Speech Recognition

2024-04-04 · Hainan Xu, Zhehuai Chen, Fei Jia, Boris Ginsburg

This paper proposes Transducers with Pronunciation-aware Embeddings (PET). Unlike conventional Transducers where the decoder embeddings for different tokens are trained independently, the PET model's decoder embedding in…

Automatic Speech RecognitionDecoderspeech-recognitionSpeech Recognition

TTS-Transducer: End-to-End Speech Synthesis with Neural Transducer

2025-01-10 · Vladimir Bataev, Subhankar Ghosh, Vitaly Lavrukhin, Jason Li

This work introduces TTS-Transducer - a novel architecture for text-to-speech, leveraging the strengths of audio codec models and neural transducers. Transducers, renowned for their superior quality and robustness in spe…

speech-recognitionSpeech RecognitionSpeech Synthesistext-to-speech+1

Large-Scale Streaming End-to-End Speech Translation with Neural Transducers

2022-04-11 · Jian Xue, Peidong Wang, Jinyu Li, Matt Post 외

Neural transducers have been widely used in automatic speech recognition (ASR). In this paper, we introduce it to streaming end-to-end speech translation (ST), which aims to convert audio signals to texts in other langua…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Machine Translationspeech-recognition+2

Neural Finite-State Transducers: Beyond Rational Relations

2019-06-01 · NAACL 2019 6 · Chu-Cheng Lin, Hao Zhu, Matthew R. Gormley, Jason Eisner

We introduce neural finite state transducers (NFSTs), a family of string transduction models defining joint and conditional probability distributions over pairs of strings. The probability of a string pair is obtained by…

Neural Induction of Finite-State Transducers

2026-01-16 · Michael Ginn, Alexis Palmer, Mans Hulden arxiv

Finite-State Transducers (FSTs) are effective models for string-to-string rewriting tasks, often providing the efficiency necessary for high-performance applications, but constructing transducers by hand is difficult. In…