About the job
About Valka
Valka, an innovative spin-off from the Realms Group (parent company of Oddin.gg), aims to transform how individuals create and engage with digital content. Our vision is to foster a platform where content is not merely consumed but co-created in real-time, merging imagination with reality. With the power of advanced AI, we strive to develop an interactive human-digital platform that allows virtual characters to react dynamically to users through voice, text, gestures, and more.
This is an opportunity to be part of a diverse team of innovators dedicated to pushing the boundaries of generative content. Our mission is to shift from passive consumption to active participation, opening new creative avenues across gaming, entertainment, education, and beyond.
Position Overview:
We are seeking a skilled Applied Research Scientist in Speech Synthesis to join our pioneering team. In this crucial role, you will develop text-to-speech and voice cloning models designed to produce synthetic voices that emulate public figures.
You will work with cutting-edge models to push the boundaries of voice cloning and TTS technology. A deep understanding of speech synthesis, natural language processing (NLP), and deep learning is essential, along with experience in handling large text and speech datasets.
Your responsibilities will include creating efficient training and deployment pipelines for voice models. You will also design validation strategies to compare synthetic speech with real recordings and develop custom metrics to assess quality.
Additionally, you will assist in establishing the infrastructure for tracking experiments, ensuring reproducibility of results, and deploying models in production. Your involvement will span the complete machine learning workflow, from training on distributed systems to monitoring deployed models.
