1 month ago
Sŏngnam, Korea, SouthMid Level
Responsibilities
- Optimize TTS models for on-device and server environments.
- Port and optimize models for NPU and other AI accelerator environments.
- Lighten models through quantization and optimize memory, computation, and performance.
- Develop real-time streaming speech synthesis systems and optimize latency.
- Develop and deploy multilingual and multi-speaker TTS models.
- Research and develop modern TTS models based on LLMs and Flow Matching.
- Build and improve speech-synthesis datasets based on modern generative models.
- Analyze and resolve TTS model and service issues in production environments.
Requirements
- At least 3 years of relevant TTS experience.
- Foundational knowledge of speech synthesis and TTS concepts.
- Experience with model compression and computational optimization, including quantization, memory, and latency optimization.
- Experience porting and optimizing models for NPUs or other AI accelerator environments.
- Experience optimizing models with TensorRT and ONNX.
- Understanding of audio signal processing.
- Development and debugging experience in Linux environments.
- Programming ability in Python, C++, shell, or similar languages.
- Ability to use open-source deep-learning frameworks such as PyTorch.
- Preferred: multilingual or multi-speaker TTS development experience; understanding or experience with speech/audio tokenization; publications in leading speech synthesis, machine learning, or AI venues; or a master's/doctoral degree in a related field.
Benefits
- A 3-month probationary period may apply.
- The interview process may include document screening, a video interview, an in-person or video interview, and offer discussions; steps may change depending on schedule and progress.
- Veterans and employment-protected candidates are given preference according to applicable law.
- Registered persons with disabilities are given preference according to applicable law.
- A reference check may be conducted with the applicant's consent after the interview process.
Categories
About 42dot
42dot builds software-defined vehicle and autonomous mobility platforms for automakers and mobility service providers. Its products include AI-driven vehicle software, data-based update services, and core platform components such as API gateways and identity/access management to connect services securely. The privately held company is headquartered in Seongnam-si, Gyeonggi-do, and focuses on transportation and automotive markets.