Cover Image

Model-free Speculative Decoding for Transformer-based ASR with Token Map Drafting

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Ho, Tuan Vu, Kokubo, Hiroaki, Yamamoto, Masaaki, Kawaguchi, Yohei
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Sound Audio and Speech Processing
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

LLM-based Generative Error Correction for Rare Words with Synthetic Data and Phonetic Context
by: Yamashita, Natsuo, et al.
Published: (2025)

Synthetic Data Domain Adaptation for ASR via LLM-based Text and Phonetic Respelling Augmentation
by: Yamashita, Natsuo, et al.
Published: (2026)

Stream-based Active Learning for Anomalous Sound Detection in Machine Condition Monitoring
by: Ho, Tuan Vu, et al.
Published: (2024)

End-to-End Integration of Speech Emotion Recognition with Voice Activity Detection using Self-Supervised Learning Features
by: Yamashita, Natsuo, et al.
Published: (2024)

Loss Masking Is Not Needed in Decoder-only Transformer for Discrete-token-based ASR
by: Chen, Qian, et al.
Published: (2023)

Exploring SSL Discrete Tokens for Multilingual ASR
by: Cui, Mingyu, et al.
Published: (2024)

AsyncSwitch: Asynchronous Text-Speech Adaptation for Code-Switched ASR
by: Nguyen, Tuan, et al.
Published: (2025)

TokenVerse: Towards Unifying Speech and NLP Tasks via Transducer-based ASR
by: Kumar, Shashi, et al.
Published: (2024)

Diagnostic-Driven Layer-Wise Compensation for Post-Training Quantization of Encoder-Decoder ASR Models
by: Wang, Xinyu, et al.
Published: (2026)

Continual Learning Optimizations for Auto-regressive Decoder of Multilingual ASR systems
by: Kwok, Chin Yuen, et al.
Published: (2024)

SpecASR: Accelerating LLM-based Automatic Speech Recognition via Speculative Decoding
by: Wei, Linye, et al.
Published: (2025)

Improving noisy student training for low-resource languages in End-to-End ASR using CycleGAN and inter-domain losses
by: Li, Chia-Yu, et al.
Published: (2024)

Pushing the Performance of Synthetic Speech Detection with Kolmogorov-Arnold Networks and Self-Supervised Learning Models
by: Phuong, Tuan Dat, et al.
Published: (2025)

MSA-ASR: Efficient Multilingual Speaker Attribution with frozen ASR Models
by: Nguyen, Thai-Binh, et al.
Published: (2024)

AdaCS: Adaptive Normalization for Enhanced Code-Switching ASR
by: Chu, The Chuong, et al.
Published: (2025)

WCTC-Biasing: Retraining-free Contextual Biasing ASR with Wildcard CTC-based Keyword Spotting and Inter-layer Biasing
by: Nakagome, Yu, et al.
Published: (2025)

ASR-EC Benchmark: Evaluating Large Language Models on Chinese ASR Error Correction
by: Wei, Victor Junqiu, et al.
Published: (2024)

PromptASR for contextualized ASR with controllable style
by: Yang, Xiaoyu, et al.
Published: (2023)

WhisperKit: On-device Real-time ASR with Billion-Scale Transformers
by: Orhon, Atila, et al.
Published: (2025)

Acoustic scattering AI for non-invasive object classifications: A case study on hair assessment
by: Hoang, Long-Vu, et al.
Published: (2025)

Qwen vs. Gemma Integration with Whisper: A Comparative Study in Multilingual SpeechLLM Systems
by: Nguyen, Tuan, et al.
Published: (2025)

Are Transformers in Pre-trained LM A Good ASR Encoder? An Empirical Study
by: An, Keyu, et al.
Published: (2024)

ManWav: The First Manchu ASR Model
by: Seo, Jean, et al.
Published: (2024)

Transformer-based Model for ASR N-Best Rescoring and Rewriting
by: Kang, Iwen E., et al.
Published: (2024)

Alignment-Free Training for Transducer-based Multi-Talker ASR
by: Moriya, Takafumi, et al.
Published: (2024)

Performant ASR Models for Medical Entities in Accented Speech
by: Afonja, Tejumade, et al.
Published: (2024)

ASR Error Correction using Large Language Models
by: Ma, Rao, et al.
Published: (2024)

WER We Stand: Benchmarking Urdu ASR Models
by: Arif, Samee, et al.
Published: (2024)

AutoMode-ASR: Learning to Select ASR Systems for Better Quality and Cost
by: Gündüz, Ahmet, et al.
Published: (2024)

Locality enhanced dynamic biasing and sampling strategies for contextual ASR
by: Jalal, Md Asif, et al.
Published: (2024)

Boosting Hybrid Autoregressive Transducer-based ASR with Internal Acoustic Model Training and Dual Blank Thresholding
by: Moriya, Takafumi, et al.
Published: (2024)

Exploring SSL Discrete Speech Features for Zipformer-based Contextual ASR
by: Cui, Mingyu, et al.
Published: (2024)

Romanization Encoding For Multilingual ASR
by: Ding, Wen, et al.
Published: (2024)

NIM4-ASR: Towards Efficient, Robust, and Customizable Real-Time LLM-Based ASR
by: Xie, Yuan, et al.
Published: (2026)

LA-RAG:Enhancing LLM-based ASR Accuracy with Retrieval-Augmented Generation
by: Li, Shaojun, et al.
Published: (2024)

Benchmarking Children's ASR with Supervised and Self-supervised Speech Foundation Models
by: Fan, Ruchao, et al.
Published: (2024)

Promptformer: Prompted Conformer Transducer for ASR
by: Duarte-Torres, Sergio, et al.
Published: (2024)

Qwen3-ASR Technical Report
by: Shi, Xian, et al.
Published: (2026)

Revisiting Acoustic Features for Robust ASR
by: Shah, Muhammad A., et al.
Published: (2024)

Effective Text Adaptation for LLM-based ASR through Soft Prompt Fine-Tuning
by: Ma, Yingyi, et al.
Published: (2024)