paper-with-me

Papers

SpeechLM: Enhanced Speech Pre-Training with Unpaired Textual Data

2022-09-30 · Ziqiang Zhang, Sanyuan Chen, Long Zhou, Yu Wu, Shuo Ren, Shujie Liu, Zhuoyuan Yao, Xun Gong, LiRong Dai, Jinyu Li, Furu Wei

How to boost speech pre-training with textual data is an unsolved problem due to the fact that speech and text are very different modalities with distinct characteristics. In this paper, we propose a cross-modal Speech and Language Model (SpeechLM) to explicitly align speech and text pre-training with a pre-defined unified discrete representation. Specifically, we introduce two alternative discrete tokenizers to bridge the speech and text modalities, including phoneme-unit and hidden-unit tokenizers, which can be trained using a small amount of paired speech-text data. Based on the trained tokenizers, we convert the unlabeled speech and text data into tokens of phoneme units or hidden units. The pre-training objective is designed to unify the speech and the text into the same discrete semantic space with a unified Transformer network. We evaluate SpeechLM on various spoken language processing tasks including speech recognition, speech translation, and universal representation evaluation framework SUPERB, demonstrating significant improvements on content-related tasks. Code and models are available at https://aka.ms/SpeechLM.

📄 PDF Abstract BibTeX arXiv:2209.15329

Code (1)

microsoft/speecht5 pytorch

Tasks

Language ModelingLanguage Modellingspeech-recognitionSpeech Recognition

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Multi-Head Attention 설명 없음
Adam 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음

Similar Papers 제목 키워드 기반

Textually Pretrained Speech Language Models

2023-05-22 · NeurIPS 2023 11 · Michael Hassid, Tal Remez, Tu Anh Nguyen, Itai Gat 외

Speech language models (SpeechLMs) process and generate acoustic data only, without textual supervision. In this work, we propose TWIST, a method for training SpeechLMs using a warm-start from a pretrained textual langua…

Language Modelling

ESPnet-SpeechLM: An Open Speech Language Model Toolkit

2025-02-21 · Jinchuan Tian, Jiatong Shi, William Chen, Siddhant Arora 외

We present ESPnet-SpeechLM, an open toolkit designed to democratize the development of speech language models (SpeechLMs) and voice-driven agentic applications. The toolkit standardizes speech processing tasks by framing…

Language ModelingLanguage Modelling

Enhancing Speech Large Language Models through Reinforced Behavior Alignment

2025-08-25 · Yansong Liu, Jiateng Li, Yuan Liu arxiv

The recent advancements of Large Language Models (LLMs) have spurred considerable research interest in extending their linguistic capabilities beyond text to other modalities, which leads to emergence of speech-based LLM…

Speech-to-Text TranslationReinforcement LearningQuestion Answering

OpusLM: A Family of Open Unified Speech Language Models

2025-06-21 · Jinchuan Tian, William Chen, Yifan Peng, Jiatong Shi 외

This paper presents Open Unified Speech Language Models (OpusLMs), a family of open foundational speech language models (SpeechLMs) up to 7B. Initialized from decoder-only text language models, the OpusLMs are continuous…

Decoderspeech-recognitionSpeech RecognitionSpeech Synthesis

SpeechMedAssist: Efficiently and Effectively Adapting Speech Language Models for Medical Consultation

2026-01-08 · Sirry Chen, Jieyi Wang, Wei Chen, Zhongyu Wei arxiv

Medical consultations are intrinsically speech-centric. However, most prior works focus on long-text-based interactions, which are cumbersome and patient-unfriendly. Recent advances in speech language models (SpeechLMs) …

Question Answering