Training dynamic models using early exits for automatic speech recognition on resource-constrained devices

Documenti analoghi

• Splitformer: An improved early-exit architecture for automatic speech recognition on edge devices
  di: Lasbordes, Maxence, et al.
  Pubblicazione: (2025)
• Efficient Fine-tuning of Audio Spectrogram Transformers via Soft Mixture of Adapters
  di: Cappellazzo, Umberto, et al.
  Pubblicazione: (2024)
• Parameter-Efficient Transfer Learning of Audio Spectrogram Transformers
  di: Cappellazzo, Umberto, et al.
  Pubblicazione: (2023)
• Continual Contrastive Spoken Language Understanding
  di: Cappellazzo, Umberto, et al.
  Pubblicazione: (2023)
• Scaling and Enhancing LLM-based AVSR: A Sparse Mixture of Projectors Approach
  di: Cappellazzo, Umberto, et al.
  Pubblicazione: (2025)