English
Related papers

Related papers: Bytes are All You Need: End-to-End Multilingual Sp…

200 papers

Subword tokenization is an essential part of modern large language models (LLMs), yet its specific contributions to training efficiency and model performance remain poorly understood. In this work, we decouple the effects of subword…

Computation and Language · Computer Science 2026-05-15 Théo Gigant , Bowen Peng , Jeffrey Quesnelle

A Spoken dialogue system for an unseen language is referred to as Zero resource speech. It is especially beneficial for developing applications for languages that have low digital resources. Zero resource speech synthesis is the task of…

Audio and Speech Processing · Electrical Eng. & Systems 2020-09-11 Karthik Pandia D S , Anusha Prakash , Mano Ranjith Kumar , Hema A Murthy

End-to-end speech-to-text translation models are often initialized with pre-trained speech encoder and pre-trained text decoder. This leads to a significant training gap between pre-training and fine-tuning, largely due to the modality…

Computation and Language · Computer Science 2022-07-05 Jinming Zhao , Hao Yang , Ehsan Shareghi , Gholamreza Haffari

Subword tokenization is a commonly used input pre-processing step in most recent NLP models. However, it limits the models' ability to leverage end-to-end task learning. Its frequency-based vocabulary creation compromises tokenization in…

Computation and Language · Computer Science 2022-04-25 Md Mofijul Islam , Gustavo Aguilar , Pragaash Ponnusamy , Clint Solomon Mathialagan , Chengyuan Ma , Chenlei Guo

While most deployed speech recognition systems today still run on servers, we are in the midst of a transition towards deployments on edge devices. This leap to the edge is powered by the progression from traditional speech recognition…

Computation and Language · Computer Science 2020-02-10 Yuan Shangguan , Jian Li , Qiao Liang , Raziel Alvarez , Ian McGraw

End-to-end architectures have been recently proposed for spoken language understanding (SLU) and semantic parsing. Based on a large amount of data, those models learn jointly acoustic and linguistic-sequential features. Such architectures…

Computation and Language · Computer Science 2020-02-17 Marco Dinarelli , Nikita Kapoor , Bassam Jabaian , Laurent Besacier

Audio grounding, or speech-driven open-set object detection, aims to localize and identify objects directly from speech, enabling generalization beyond predefined categories. This task is crucial for applications like human-robot…

Sound · Computer Science 2025-09-23 Wenhuan Lu , Xinyue Song , Wenjun Ke , Zhizhi Yu , Wenhao Yang , Jianguo Wei

Pretrained character-level and byte-level language models have been shown to be competitive with popular subword models across a range of Natural Language Processing (NLP) tasks. However, there has been little research on their…

Computation and Language · Computer Science 2024-05-24 Lukas Edman , Gabriele Sarti , Antonio Toral , Gertjan van Noord , Arianna Bisazza

Multi-speaker speech recognition has been one of the keychallenges in conversation transcription as it breaks the singleactive speaker assumption employed by most state-of-the-artspeech recognition systems. Speech separation is consideredas…

Audio and Speech Processing · Electrical Eng. & Systems 2020-09-08 Jian Wu , Zhuo Chen , Jinyu Li , Takuya Yoshioka , Zhili Tan , Ed Lin , Yi Luo , Lei Xie

With the similarity between music and speech synthesis from symbolic input and the rapid development of text-to-speech (TTS) techniques, it is worthwhile to explore ways to improve the MIDI-to-audio performance by borrowing from TTS…

Sound · Computer Science 2023-03-22 Xuan Shi , Erica Cooper , Xin Wang , Junichi Yamagishi , Shrikanth Narayanan

Tokenization is the first -- and often least scrutinized -- step of most NLP pipelines. Standard algorithms for learning tokenizers rely on frequency-based objectives, which favor languages dominant in the training data and consequently…

Computation and Language · Computer Science 2025-08-25 Negar Foroutan , Clara Meister , Debjit Paul , Joel Niklaus , Sina Ahmadi , Antoine Bosselut , Rico Sennrich

End-to-end (E2E) systems are fast replacing the conventional systems in the domain of automatic speech recognition. As the target labels are learned directly from speech data, the E2E systems need a bigger corpus for effective training. In…

Audio and Speech Processing · Electrical Eng. & Systems 2019-07-22 Kunal Dhawan , Ganji Sreeram , Kumar Priyadarshi , Rohit Sinha

Currently, high-quality, synchronized audio is synthesized using various multi-modal joint learning frameworks, leveraging video and optional text inputs. In the video-to-audio benchmarks, video-to-audio quality, semantic alignment, and…

Computer Vision and Pattern Recognition · Computer Science 2025-03-31 Haomin Zhang , Chang Liu , Junjie Zheng , Zihao Chen , Chaofan Ding , Xinhan Di

End-to-end text-to-speech (TTS) has shown great success on large quantities of paired text plus speech data. However, laborious data collection remains difficult for at least 95% of the languages over the world, which hinders the…

Computation and Language · Computer Science 2019-07-03 Tao Tu , Yuan-Jui Chen , Cheng-chieh Yeh , Hung-yi Lee

Single-stage text-to-speech models have been actively studied recently, and their results have outperformed two-stage pipeline systems. Although the previous single-stage model has made great progress, there is room for improvement in terms…

Sound · Computer Science 2023-08-01 Jungil Kong , Jihoon Park , Beomjeong Kim , Jeongmin Kim , Dohee Kong , Sangjin Kim

Recent studies have introduced end-to-end TTS, which integrates the production of context and acoustic features in statistical parametric speech synthesis. As a result, a single neural network replaced laborious feature engineering with…

Machine Learning · Computer Science 2019-02-26 Kohki Mametani , Tsuneo Kato , Seiichi Yamamoto

Neural sequence-to-sequence text-to-speech synthesis (TTS) can produce high-quality speech directly from text or simple linguistic features such as phonemes. Unlike traditional pipeline TTS, the neural sequence-to-sequence TTS does not…

Audio and Speech Processing · Electrical Eng. & Systems 2020-10-08 Yusuke Yasuda , Xin Wang , Junichi Yamagishi

End-to-end (E2E) speech-to-text translation (ST) often depends on pretraining its encoder and/or decoder using source transcripts via speech recognition or text translation tasks, without which translation performance drops substantially.…

Computation and Language · Computer Science 2022-06-10 Biao Zhang , Barry Haddow , Rico Sennrich

All-neural end-to-end (E2E) automatic speech recognition (ASR) systems that use a single neural network to transduce audio to word sequences have been shown to achieve state-of-the-art results on several tasks. In this work, we examine the…

Audio and Speech Processing · Electrical Eng. & Systems 2019-10-28 Arun Narayanan , Rohit Prabhavalkar , Chung-Cheng Chiu , David Rybach , Tara N. Sainath , Trevor Strohman

Speech processing requires very efficient methods and algorithms. Finite-state transducers have been shown recently both to constitute a very useful abstract model and to lead to highly efficient time and space algorithms in this field. We…

cmp-lg · Computer Science 2008-02-03 Mehryar Mohri , Michael Riley , Richard Sproat