Open Source
Frontend-Assisted Large Audio-Language Models for SLT2026 SmartGlasses Challenge
arXiv cs.SD0 views
arXiv:2610.13282v1 Announce Type: cross Abstract: This paper describes our systems for the Track 1 time-stamped speaker-attributed ASR task of the SmartGlasses Challenge. To handle overlapping speech, background noise and far-field conditions in two-person conversations, we employ SoulX-Transcriber, a multimodal large language model that takes full-length audio as input and generates structured transcriptions. In addition, we investigate the effect of speech frontend processing, including a rule-based Guided Source Separation method and a data-driven speech separation model, both of which improve performance on the validation set. Considering inference stability, our final submission uses the fine-tuned SoulX-Transcriber system, achieving a tcpCER of 6.22 on the test set.
Source：https://arxiv.org/abs/2610.13282PreviousDoes Location Matter? A Black-Box Causal Audit Reveals Model-Dependent Audio Grounding in LALM HallucinationNextSonic-AMG: Streaming Human Motion Generation from Spatial Audio
Related posts
Your Product Is Copyable. Your Distribution Isn't: A 24-Channel Map for Solo Builders
0dev.to
Astra can read and write weights to improve Qwen
0Hacker News
Distillation didn't help our small model until we took the labels away
0dev.to Machine Learning
462. Minimum Moves to Equal Array Elements II
0dev.to PHP
MUSBRIC: LEGO-Compatible Pneumatic Muscle Bricks for Reconfigurable Musculoskeletal Robots
0arXiv cs.RO
OpenFlyV2: An onboard vision-language-action model for aerial perception and navigation
0arXiv cs.RO