🛠 Tech Stack

💼 About This Role

You'll research and train state-of-the-art TTS models for realistic emotional voice generation in entertainment and education. Your work will directly power real-time interactive virtual characters on a cutting-edge AI platform. This role offers the chance to redefine digital content creation at a visionary startup.

🎯 What You'll Do

• Maintain and improve the current best TTS model pipeline.
• Assemble results from experiments into a production speech system.
• Evaluate model performance and quality metrics.
• Collaborate with TTS researchers and product engineering teams.

📋 Requirements

• Experience with ML models in production (formats, quantization, deployment).
• Proficiency in Python and key libraries (PyTorch, Hugging Face Transformers).
• Knowledge of transformers, diffusion models, or GANs.

✨ Nice to Have

• Experience with text-to-speech or voice cloning model training.
• Familiarity with modern speech synthesis models (GPT-based, flow matching).
• Contributions to open-source AI tools.

🎁 Benefits & Perks

• 🚀 Cutting-edge project at an innovative AI startup.
• 🌍 Remote work with flexible hours.
• 🤝 Collaborative team of researchers and engineers.