Gil Keren

Computer Science · Physical Sciences

13h-index663citations54works0.32yr avg
OpenAlex↗ORCID↗Google Scholar↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Speech Recognition and Synthesis(28), Speech and Audio Processing(21), Music and Audio Processing(17), Natural Language Processing Techniques(9), Domain Adaptation and Few-Shot Learning(9)

Publications53 total

Omnilingual ASR: Open-Source Multilingual Speech Recognition for 1600+ Languages
arXiv (Cornell University)·2025Open Access
Faster Speech-LLaMA Inference with Multi-token Prediction
2025· 2 cited
Efficient Streaming LLM for Speech Recognition
2025
M-BEST-RQ: A Multi-Channel Speech Foundation Model for Smart Glasses
2025
Speech-N-LlaMA: Improving Speech LLMs with Multi-Pass Training
2025
Token-Weighted RNN-T For Learning From Flawed Data
2024
Efficient Streaming LLM for Speech Recognition
arXiv (Cornell University)·2024Open Access
M-BEST-RQ: A Multi-Channel Speech Foundation Model for Smart Glasses
arXiv (Cornell University)·2024Open Access
Faster Speech-LLaMA Inference with Multi-token Prediction
arXiv (Cornell University)·2024Open Access
Token-Weighted RNN-T for Learning from Flawed Data
arXiv (Cornell University)·2024· 1 citedOpen Access
A Token-Wise Beam Search Algorithm for RNN-T
2023· 2 cited
Towards Selection of Text-to-speech Data to Augment ASR Training
arXiv (Cornell University)·2023· 1 citedOpen Access
Text Generation with Speech Synthesis for ASR Data Augmentation
arXiv (Cornell University)·2023· 1 citedOpen Access
Improving fast-slow Encoder based Transducer with Streaming Deliberation
2023· 4 cited
A Token-Wise Beam Search Algorithm for RNN-T
arXiv (Cornell University)·2023Open Access
PersonaLM: Language Model Personalization via Domain-distributed Span Aggregated K-Nearest N-gram Retrieval Augmentation
2023· 1 citedOpen Access
Improving Fast-slow Encoder based Transducer with Streaming Deliberation
arXiv (Cornell University)·2022Open Access
Scaling ASR Improves Zero and Few Shot Learning
Interspeech 2022·2022· 8 cited
Scaling ASR Improves Zero and Few Shot Learning
arXiv (Cornell University)·2021Open Access
Contextualized Streaming End-to-End Speech Recognition with Trie-Based Deep Biasing and Shallow Fusion
2021· 59 cited
page 1 of 3Next →

Frequent Co-authors

Björn W. Schuller(25), Ozlem Kalinli(16), Jay Mahadeokar(15), Christian Fuegen(8), Junteng Jia(7), Michael L. Seltzer(6), Sivan Sabato(6), Duc Le(5), Yatharth Saraf(5), Nicholas Cummins(5), Desh Raj(5), Julian Chan(4), Yangyang Shi(4), Maximilian Schmitt(4), Manh Duc Le(4), Egor Lakomkin(4), Mahaveer Jain(3), Yuan Shangguan(3), Shahin Amiriparian(3), Ke Li(3)