paper-with-me

Papers

Optimizing Byte-level Representation for End-to-end ASR

2024-06-14 · Roger Hsiao, Liuhui Deng, Erik McDermott, Ruchir Travadi, Xiaodan Zhuang

We propose a novel approach to optimizing a byte-level representation for end-to-end automatic speech recognition (ASR). Byte-level representation is often used by large scale multilingual ASR systems when the character set of the supported languages is large. The compactness and universality of byte-level representation allow the ASR models to use smaller output vocabularies and therefore, provide more flexibility. UTF-8 is a commonly used byte-level representation for multilingual ASR, but it is not designed to optimize machine learning tasks directly. By using auto-encoder and vector quantization, we show that we can optimize a byte-level representation for ASR and achieve better accuracy. Our proposed framework can incorporate information from different modalities, and provides an error correction mechanism. In an English/Mandarin dictation task, we show that a bilingual ASR model built with this approach can outperform UTF-8 representation by 5% relative in error rate.

📄 PDF Abstract BibTeX arXiv:2406.09676

Code (0)

등록된 구현이 없습니다.

Tasks

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

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

byteSteady: Fast Classification Using Byte-Level n-Gram Embeddings

2021-06-24 · Xiang Zhang, Alexandre Drouin, Raymond Li

This article introduces byteSteady -- a fast model for classification using byte-level n-gram embeddings. byteSteady assumes that each input comes as a sequence of bytes. A representation vector is produced using the ave…

Classificationtext-classificationText Classification

Bilingual End-to-End ASR with Byte-Level Subwords

2022-05-01 · Liuhui Deng, Roger Hsiao, Arnab Ghoshal

In this paper, we investigate how the output representation of an end-to-end neural network affects multilingual automatic speech recognition (ASR). We study different representations including character-level, byte-leve…

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

Distilling Token-Trained Models into Byte-Level Models

2026-02-01 · Zishuo Bao, Jiaqi Leng, Junxiong Wang, Bowen Peng 외 arxiv

Byte Language Models (BLMs) have emerged as a promising direction for scaling language models beyond tokenization. However, existing BLMs typically require training from scratch on trillions of bytes, making them prohibi…

Knowledge Distillation

ByteFlow: Language Modeling through Adaptive Byte Compression without a Tokenizer

2026-03-03 · Chunyuan Deng, Sanket Lokegaonkar, Colin Lockard, Besnik Fetahu 외 arxiv

Modern language models still rely on fixed, pre-defined subword tokenizations. Once a tokenizer is trained, the LM can only operate at this fixed level of granularity, which often leads to brittle and counterintuitive be…

Neural Machine Translation without Embeddings

2020-08-21 · NAACL 2021 4 · Uri Shaham, Omer Levy

Many NLP models operate over sequences of subword tokens produced by hand-crafted tokenization rules and heuristic subword induction algorithms. A simple universal alternative is to represent every computerized text as a…

DecoderMachine TranslationTranslation