Deep Learning Engineer (음성 합성 개발)

The role

This person will research and develop next-generation Text-to-Speech (TTS) technology for various vehicle environments. They will work on the entire TTS development process including data construction, model training, inference optimization, and service application.

What you will do

• Optimize TTS models for On-device and Server environments
• Perform model lightweighting and performance optimization based on Quantization
• Develop real-time (streaming) speech synthesis systems and optimize latency
• Develop and apply multilingual and multi-speaker TTS models to services
• Research and develop latest TTS models based on LLM and Flow Matching
• Construct speech synthesis data and improve quality based on latest generative models
• Analyze and resolve TTS model and service issues occurring in mass production environments

Required qualifications

• 3 or more years of relevant TTS experience
• Basic concepts and knowledge of speech synthesis (TTS)
• Experience in model lightweighting and computation optimization
• Experience in Quantization, Memory / Latency optimization
• Experience in model optimization based on TensorRT, ONNX
• Understanding of Audio Signal Processing
• Experience in development and debugging in Linux environments
• Programming skills for problem solving (Python/C++/shell, etc.)
• Ability to use open-source deep learning frameworks (PyTorch, etc.)

Nice to have

• Flexibility to respond to new technologies and environmental changes and fast learning/application
• Enjoyment of directly experimenting with new AI technologies, analyzing results, and continuous improvement
• Hands-on experience in porting models to actual test equipment and analyzing execution results and performance
• Responsibility to proactively analyze and solve problems occurring in mass production environments
• Proactive challenge to uncertain problems and finding/executing solutions independently
• Experience in developing multilingual and multi-speaker TTS models
• Understanding of speech/audio tokenization technology and related research or development experience
• Author of papers in top-tier conferences and journals in speech synthesis/machine learning/AI fields
• Master's or Doctoral degree holders or those who have completed such degrees in related fields

Before you apply

• Interview process includes document screening, 1st interview (video, approx. 1 hour), 2nd interview (in-person or video, approx. 3 hours), and salary negotiation/hiring
• Do not include prohibited information such as resident registration number, family relations, marital status, salary, photo, physical conditions, or hometown in the application
• Employment may be canceled if false information is found in the application
• Reference checks may be conducted after the interview process with the candidate's consent

About the company

Gleo AI researches next-generation Text-to-Speech (TTS) technology to provide users with natural and rich voice experiences in various vehicle environments. 42dot connects research to actual products and services.