GDC is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

View, browse and sort the ever-growing list of sessions by day, pass type, topic, and format. With this Session Viewer, you can view GDC 2023 session details and speakers, and share your favorites via social media. You will be able to build your schedule and access it during the show via export or Mobile App, once live. Sessions do fill up and seating is first come, first serve, so arrive early to sessions that you would like to attend.

VoRACE: Tools for Automating Dialogue Processing

Jean-Philippe Letendre  (R&D Programmer, Ubisoft)

Location: Room 3006, West Hall

Date: Thursday, March 23

Time: 2:00 pm - 2:30 pm

Pass Type: All Access Pass, Core Pass, Audio Pass

Topic: Audio, Programming

Format: Lecture

Vault Recording: Video

Audience Level: All

Every year, AAA game companies record hundreds of thousands of dialog lines that bring to life the worlds in their games. Processing these lines is a lengthy process that requires a colossal amount of manual work. VoRACE, or Voice Recording Automating Cleaning Engine, is a set of tools using deep learning to automate the tedious processing of recordings. It accelerates the overall workflow while maintaining the same level of quality.

This lecture describes the steps involved in dialogue post-processing along with their challenges and why existing solutions do not apply. It shows how VoRACE automates the various steps in the post-processing pipeline and why it is conceived as computer-assisted creator tools rather than blunt automation modules. Lastly, it presents the implementation, deployment and testing strategy used, and reports the performance of VoRACE in a real-world production setting.

Takeaway

Attendees will learn about a novel approach to dialogue post-processing, that leverages deep learning to create a computer-assisted dialogue processing tool.

Intended Audience

This is for sound engineers, developers, and managers, but technical details will be kept to a minimum to ease comprehension for people outside the field.