Sr. Principal TTS Researcher(m/w/x)

Description

In this role, you will be at the forefront of TTS innovation, developing advanced systems and optimizing models while collaborating with a global team. Your expertise in machine learning and speech processing will drive meaningful advancements in voice technology for the automotive industry.

Requirements

• •8+ years of hands-on experience in TTS system development
• •Proficiency in C/C++ and Python
• •Strong background in NLP techniques or speech signal processing
• •Experience with linguistic tools and phonetic knowledge
• •Familiarity with transformer-based language models for prosody prediction
• •Deep understanding of autoregressive and non-autoregressive acoustic models
• •Experience optimizing models via quantization, pruning, or knowledge distillation
• •Knowledge of speech codecs and real-time streaming protocols
• •Production experience with ONNX Runtime, TensorRT, or TorchScript
• •Experience with zero-shot, one-shot, or few-shot voice cloning
• •Skilled GPU/TPU cluster and grid user
• •Fluent English
• •Track record of publications in INTERSPEECH, ICASSP, or NeurIPS
• •Track working experience in world-level TTS or AI company
• •Proven ability to lead cross-functional groups
• •Basic knowledge of information security and data privacy requirements
• •Demonstrative knowledge of information security through training programs

Tasks

• •Develop TTS systems with frontend and backend components
• •Utilize C/C++ and Python for system development
• •Implement ML frameworks like PyTorch and TensorFlow
• •Apply NLP techniques and speech signal processing
• •Use linguistic tools such as Festival
• •Leverage phonetic knowledge for TTS applications
• •Employ transformer-based models for prosody prediction
• •Optimize models through quantization and pruning
• •Utilize speech codecs like Opus and MELP
• •Implement real-time streaming protocols
• •Work with ONNX Runtime, TensorRT, or TorchScript
• •Develop zero-shot, one-shot, or few-shot voice cloning systems
• •Create emotional TTS systems
• •Manage GPU/TPU clusters and grids
• •Publish research in INTERSPEECH, ICASSP, or NeurIPS
• •Lead cross-functional teams to deliver TTS solutions

Tools & Technologies