Latest AI JobsBrowse Jobs
4

Deep Learning Engineer (음성 합성 개발)

42dotPangyo (Software Dream Center), South KoreaENGINEERING
Apply on company site
Salary
Not listed
Job type
Full-time
Work mode
Remote
Experience
Senior · 3+ yrs
Posted
2 Oct 2026
Status
Open · checked 7d ago

About this role

This role suits professionals who enjoy tackling complex technical problems and proactively experimenting with new technologies in speech generation. The Deep Learning Engineer position is located in Pangyo, South Korea, at 42dot, focusing on next-generation text-to-speech technology for vehicle environments.

Role Overview

42dot seeks a Deep Learning Engineer for the Gleo AI team in Pangyo, South Korea, to research and implement advanced text-to-speech technologies. The position involves working across the full development lifecycle, from data building and model training to inference optimization and production deployment, creating natural voice experiences for future mobility environments.

Key Responsibilities

Optimize text-to-speech models for on-device and server environments
Port and optimize models for AI accelerator environments including NPU
Implement quantization-based model compression and performance optimization
Develop streaming voice synthesis systems and reduce latency
Build and deploy multilingual and multi-speaker text-to-speech models
Research and develop advanced text-to-speech models using large language models and flow matching
Build datasets and improve quality for generative model-based voice synthesis
Analyze and resolve issues related to models and services in mass-production environments

Requirements

Minimum of 3 years of relevant experience in text-to-speech
Fundamental concepts and knowledge in speech synthesis
Experience with model lightweighting and computation optimization
Familiarity with quantization, memory optimization, and latency reduction
Experience porting and optimizing models in AI accelerator settings such as NPU
Optimization background using TensorRT and ONNX
Understanding of audio signal processing
Development and debugging experience in a Linux environment
Programming proficiency in Python, C++, or shell scripting
Ability to utilize open-source deep learning frameworks such as PyTorch

Nice to Have

Experience developing multilingual and multi-speaker text-to-speech models
Understanding and research or development experience in voice and audio tokenization technologies
Authored papers in top-tier conferences or journals within speech synthesis, machine learning, or artificial intelligence
Master's or doctoral degree or coursework completion in relevant fields
Hands-on experience testing, executing, and analyzing model performance on actual test hardware

Skills and Tools

Python
C++
Shell
PyTorch
TensorRT
ONNX
Linux
NPU and AI Accelerators

About the Company

42dot operates the Gleo AI team, which focuses on developing natural and rich in-vehicle voice experiences through advanced text-to-speech research, connecting experimental technologies directly into real products and services.

How to Apply

Interested candidates should submit their application through the official 42dot careers page.

Skills and tech stack

PythonC++ShellPyTorchTensorRTONNXLinuxNPULLM

Questions candidates ask

What is the location of the Deep Learning Engineer role?

The position is based in Pangyo at the Software Dream Center in South Korea.

How many years of experience are required?

Candidates must have at least 3 years of relevant experience in text-to-speech.

What programming languages and frameworks are used?

The role requires programming capabilities in Python, C++, or shell scripting, along with open-source deep learning frameworks like PyTorch, TensorRT, and ONNX.

What does the interview process consist of?

The process includes an initial document screening, a first round video interview lasting around one hour, and a second round in-person or video interview lasting about three hours, followed by compensation discussions.

Ready to apply?

The application is on the official 42dot careers page.

Apply on company site