Most learning-based speech enhancement pipelines depend on paired clean–noisy recordings, which are expensive or impossible to collect at scale in real-world conditions. Unsupervised routes like ...

Canary-1b-v2: Multilingual ASR + Translation (En ↔ 24 Languages) Canary-1b-v2 is a billion-parameter Encoder-Decoder model trained on Granary, delivering high-quality transcription and translation ...

Recently I was trying to deploy dolphin by vllm and after taking a look at the vllm deploy support, I installed vllm-dolphin. But after starting model by vllm using instruction "python -m ...

Microsoft is laying the groundwork for Windows 11 to morph into a genAI-driven OS. The company on Monday announced a critical AI technology that will make it possible to run generative AI (genAI) ...

Thanks for sharing this clean codebase to your cool paper and congrats to achieving sota. I got a general question which I did not understand from the paper and this codebase about the architecture ...

Abstract: The attention-based encoder-decoder (AED) speech recognition model has been widely successful in recent years. However, the joint optimization of acoustic model and language model in ...

Dialog response generation in open domain is an important research topic where the main challenge is to generate relevant and diverse responses. In this paper, we propose a new dialog pre-training ...