Voice Model Operator
Respeecher is a VC-backed startup creating a system that lets one voice sound like another, focused on very high-quality output and usability for demanding applications like TV and movies, call centers, and gaming. We have projects underway with top Hollywood studios and major video game companies. We are graduates of the Techstars startup accelerator. We really enjoy the stuff we do.
We are looking for someone to help us prepare customer datasets, train machine learning models, and post-process the results. Our main office is in Kyiv, Ukraine, but we are open to remote work.
Please peruse our job description below, and if you think you could fit into our team do not hesitate to get in touch!
We expect you to:
Have basic knowledge of Linux and comfort with the command-line
Work well both in a team and independently
Be comfortable sometimes spending hours listening to recordings
Have good English, especially listening skills
Have a good ear for subtle differences in audio quality
Would be a plus:
Knowledge of python
Knowledge of shell scripting
Knowledge of audio software like Audacity, Reaper
Interested in Max/MSP or Pure Data
Experience in audio plugin development
Duties:
Develop, maintain, and run scripts to prepare datasets and train machine learning models (bash and python)
Carefully listen to raw data and the outputs of our models, identify audio problems, and choose the best results
Post-process synthesized audio
Experiment with audio processing pipelines, dataset design, and model configuration
Benefits:
Be part of a small team working on an exciting project in a dynamic startup
Shape bleeding-edge machine learning research into a killer product
Work on major entertainment projects such as Hollywood movies and AAA video games
Inna
Схожі вакансії
2D-художник
Увійдіть або зареєструйтесь, щоб позначити вакансію як нецікаву
