Recently, the Doubao Speech Team rolled out an automated "AI Multi - Person Audio Drama" solution. This innovative solution is capable of supporting multi - role, highly expressive Text - to - Speech (TTS) performances. Moreover, it enables fully automated AI post - production processes.

In the context of cultural production, transforming novel text into high - quality multi - person audio drama finished products used to be a time - consuming and labor - intensive task. However, with this new solution, the entire process, from the initial novel text to the final high - quality multi - person audio drama, is completed end - to - end by AI. The AI takes charge of multiple aspects, including generating intelligent multi - role audio readings, adding sound effects, incorporating music and special effects, as well as performing intelligent mixing.

Built upon the newly upgraded multi - role Seed - TTS - 2.0 model and combined with AI - automated technologies for music, sound effects, special effects, and intelligent mixing, this solution delivers listening effects that are comparable to those of industry - leading human - voiced audio dramas. At the same time, it significantly cuts down on production costs and shortens the turnaround times for audiobook creation, which aligns with the fast - paced and cost - efficient demands of the modern cultural and entertainment industry.