13 papers
Mind the Accent Gap: British Accent Robustness in Speech-Driven Financial Voice Assistants
Aadam Haq, Oggi Rudovic, Malcolm Chadwick +5
AI voice assistants often use Automatic Speech Recognition (ASR) with LLM-based reasoning, yet existing systems struggle with regional British accents, including Scottish, Irish, a…
FraQAT: Quantization Aware Training with Fractional bits
Luca Morreale, Alberto Gil C. P. Ramos, Malcolm Chadwick +4
State-of-the-art (SOTA) generative models have demonstrated impressive capabilities in image synthesis or text generation, often with a large capacity model. However, these large m…
Efficient High-Resolution Image Editing with Hallucination-Aware Loss and Adaptive Tiling
Young D. Kwon, Abhinav Mehrotra, Malcolm Chadwick +2
High-resolution (4K) image-to-image synthesis has become increasingly important for mobile applications. Existing diffusion models for image editing face significant challenges, in…
HierarchicalPrune: Position-Aware Compression for Large-Scale Diffusion Models
Young D. Kwon, Rui Li, Sijia Li +3
State-of-the-art text-to-image diffusion models (DMs) achieve remarkable quality, yet their massive parameter scale (8-11B) poses significant challenges for inferences on resource-…
Robust Unsupervised Adaptation of a Speech Recogniser Using Entropy Minimisation and Speaker Codes
Rogier C. van Dalen, Shucong Zhang, Titouan Parcollet +1
Speech recognisers usually perform optimally only in a specific environment and need to be adapted to work well in another. For adaptation to a new speaker, there is often too litt…
Evaluation of LLMs in Speech is Often Flawed: Test Set Contamination in Large Language Models for Speech Recognition
Yuan Tseng, Titouan Parcollet, Rogier van Dalen +2
Recent work suggests that large language models (LLMs) can improve performance of speech tasks compared to existing systems. To support their claims, results on LibriSpeech and Com…
Guidance Free Image Editing via Explicit Conditioning
Mehdi Noroozi, Alberto Gil Ramos, Luca Morreale +4
Current sampling mechanisms for conditional diffusion models rely mainly on Classifier Free Guidance (CFG) to generate high-quality images. However, CFG requires several denoising…
Upcycling Text-to-Image Diffusion Models for Multi-Task Capabilities
Ruchika Chavhan, Abhinav Mehrotra, Malcolm Chadwick +4
Text-to-image synthesis has witnessed remarkable advancements in recent years. Many attempts have been made to adopt text-to-image models to support multiple tasks. However, existi…
EDiT: Efficient Diffusion Transformers with Linear Compressed Attention
Philipp Becker, Abhinav Mehrotra, Ruchika Chavhan +5
Diffusion Transformers (DiTs) have emerged as a leading architecture for text-to-image synthesis, producing high-quality and photorealistic images. However, the quadratic scaling p…
Benchmarking Rotary Position Embeddings for Automatic Speech Recognition
Shucong Zhang, Titouan Parcollet, Rogier van Dalen +1
Self-attention relies on positional embeddings to encode input order. Relative Position (RelPos) embeddings are widely used in Automatic Speech Recognition (ASR). However, RelPos h…
MobileQuant: Mobile-friendly Quantization for On-device Language Models
Fuwen Tan, Royson Lee, Łukasz Dudziak +5
Large language models (LLMs) have revolutionized language processing, delivering outstanding results across multiple applications. However, deploying LLMs on edge devices poses sev…
Linear-Complexity Self-Supervised Learning for Speech Processing
Shucong Zhang, Titouan Parcollet, Rogier van Dalen +1
Self-supervised learning (SSL) models usually require weeks of pre-training with dozens of high-end GPUs. These models typically have a multi-headed self-attention (MHSA) context e…
Fast Sampling Through The Reuse Of Attention Maps In Diffusion Models
Rosco Hunter, Łukasz Dudziak, Mohamed S. Abdelfattah +3
Text-to-image diffusion models have demonstrated unprecedented capabilities for flexible and realistic image synthesis. Nevertheless, these models rely on a time-consuming sampling…