BLR-MoE: Boosted Language-Routing Mixture of Experts for Domain-Robust Multilingual E2E ASR

Documenti analoghi

• MoE-TTS: Enhancing Out-of-Domain Text Understanding for Description-based TTS via Mixture-of-Experts
  di: Xue, Heyang, et al.
  Pubblicazione: (2025)
• SC-MoE: Switch Conformer Mixture of Experts for Unified Streaming and Non-streaming Code-Switching ASR
  di: Ye, Shuaishuai, et al.
  Pubblicazione: (2024)
• A Language-Agnostic Hierarchical LoRA-MoE Architecture for CTC-based Multilingual ASR
  di: Zheng, Yuang, et al.
  Pubblicazione: (2026)
• On-the-fly Routing for Zero-shot MoE Speaker Adaptation of Speech Foundation Models for Dysarthric Speech Recognition
  di: HU, Shujie, et al.
  Pubblicazione: (2025)
• HDMoLE: Mixture of LoRA Experts with Hierarchical Routing and Dynamic Thresholds for Fine-Tuning LLM-based ASR Models
  di: Mu, Bingshen, et al.
  Pubblicazione: (2024)