Canary-1B-v2 & Parakeet-TDT-0.6B-v3: Efficient and High-Performance Models for Multilingual ASR and AST

Documenti analoghi

• Longer is (Not Necessarily) Stronger: Punctuated Long-Sequence Training for Enhanced Speech Recognition and Translation
  di: Koluguri, Nithin Rao, et al.
  Pubblicazione: (2024)
• Codec-ASR: Training Performant Automatic Speech Recognition Systems with Discrete Speech Representations
  di: Dhawan, Kunal, et al.
  Pubblicazione: (2024)
• BESTOW: Efficient and Streamable Speech Language Model with the Best of Two Worlds in GPT and T5
  di: Chen, Zhehuai, et al.
  Pubblicazione: (2024)
• Speaker Targeting via Self-Speaker Adaptation for Multi-talker ASR
  di: Wang, Weiqing, et al.
  Pubblicazione: (2025)
• Training and Inference Efficiency of Encoder-Decoder Speech Models
  di: Żelasko, Piotr, et al.
  Pubblicazione: (2025)