音频算法工程师-抖音直播
Core
Building low-latency audio understanding and generation models for real-time dialogue agents in Douyin live streaming, covering ASR, TTS, end-to-end speech large models, and audio classification.
Role type
Senior IC machine-learning engineer (audio)
Builds
Real-time dialogue agents for live streaming
Domain
Consumer media / Audio AI
Deliverable
production ML models
Required skills
End-to-end speech large models, ASR, TTS, audio classification, Python, C++, TensorFlow, PyTorch, Linux, data structures
Preferred skills
Publications in ICASSP, Interspeech, NIPS, ICML, ICLR, top-tier competition awards
Responsibilities
Optimize algorithms for key scenarios to build high-quality low-latency agent systems, explore boundaries of multimodal perception and interaction capabilities, and land products.
Sourced via bytedance · Listed on CareerPlan, which tracks 844,000+ jobs from 20+ sources.