Research Scientist for Speech Synthesis

📋 Description

• Design, develop, and optimize text-to-speech models preserving voice actor style
• Collaborate on audio datasets with voice recordings and multilingual transcriptions
• Experiment with neural TTS and voice cloning architectures
• Define validation strategies with metrics comparing synthetic vs real
• Build MLOps infrastructure: tracking, CI/CD, deployment, monitoring

🎯 Requirements

• PyTorch, TensorFlow, or JAX for DL
• Audio processing: sampling, spectrograms, vocoders
• Training text-to-speech and voice cloning models with Python
• Speech synthesis models: WaveNet, Tacotron, VITS
• Voice cloning models like XTTS or YourTTS
• Transformers and diffusion models