Research Scientist - Speech & Audio Understanding (Speech Generation)

Job Responsibilities:
1. Track the latest research in speech generation algorithms, explore next-generation paradigms for speech/audio generation, and push the boundaries of speech generation capabilities.  
2. Investigate cutting-edge multimodal voice foundation model technologies to enhance voice interaction experiences by integrating text, speech, and vision.  
3. Lead the technical R&D of voice foundation models, driving model performance improvements and innovative applications.  
Work Location: US-Washington-Bellevue

Job Requirements:
1. Master’s or Ph.D. in Computer Science, Artificial Intelligence, Electronic Engineering, Signal Processing, or related fields.  
2. Research or development experience in one or more areas: voice foundation models, speech synthesis, speech recognition, audio generation, voice conversion, or speech codec.  
3. Familiarity with mainstream voice-enabled large models (e.g., GPT4o, GLM-4-Voice, Qwen2.5-Omni, Voila). Prior project experience is preferred.  
4. Proficient in deep learning frameworks (e.g., PyTorch). Experience with large-scale model training frameworks (Megatron/Deepspeed) is a plus.  
5. Solid understanding of large model architectures and principles. Experience in large-scale pretraining or post-training is preferred.  。加分项：