4 papers
Sequence-Aware Split Heuristic to Mitigate SM Underutilization in FlashAttention-3 Low-Head-Count Decoding
Martí Llopart Font, Javier Hernando, Cristina España-Bonet
The standard FlashAttention-3 heuristic exhibits a GPU occupancy bottleneck in low-head-count decoding configurations because it disables sequence splitting based on sequence lengt…
Zero-Shot TTS With Enhanced Audio Prompts: Bsc Submission For The 2026 Wildspoof Challenge TTS Track
Jose Giraldo, Alex Peiró-Lilja, Rodolfo Zevallos +1
We evaluate two non-autoregressive architectures, StyleTTS2 and F5-TTS, to address the spontaneous nature of in-the-wild speech. Our models utilize flexible duration modeling to im…
Listening or Reading? Evaluating Speech Awareness in Chain-of-Thought Speech-to-Text Translation
Jacobo Romero-Díaz, Gerard I. Gállego, Oriol Pareras +3
Speech-to-Text Translation (S2TT) systems built from Automatic Speech Recognition (ASR) and Text-to-Text Translation (T2TT) modules face two major limitations: error propagation an…
Revisiting Direct Speech-to-Text Translation with Speech LLMs: Better Scaling than CoT Prompting?
Oriol Pareras, Gerard I. Gállego, Federico Costa +2
Recent work on Speech-to-Text Translation (S2TT) has focused on LLM-based models, introducing the increasingly adopted Chain-of-Thought (CoT) prompting, where the model is guided t…