Two months after the public test, Wanzhuang Yousheng officially launched its automated production system, "Fully Automatic AI Multi-voice Audiobook Creation," which had previously impressed rights holders during internal demonstrations.
This AI audio content platform, founded by the core team of Lazy Listen, completed the public test of this strategic-level feature in early June. At its core is a combination of previously validated AI capabilities—intelligent chapter splitting, character analysis, script generation, multi-character voice acting, and post-production synthesis—packaged together through a task orchestration engine to form an automated digital audio production line that can run "unmanned."
This is not an independent new product but a practical implementation of the platform's "dual-track production engine" strategy. It shares the same technical infrastructure with Wanzhuang Yousheng's existing professional creation workbench but can operate in fully automatic mode, targeting the industry's most challenging pain points: how to achieve a leap in capacity and cost reduction while maintaining quality.
From "Manual" to "Autopilot": One Engine, Two Tracks
Traditional high-quality audiobook production is a long and expensive manual process: manuscript sorting, chapter splitting, script creation, casting, voice acting, alignment, post-production mixing, listening, and export. A high-quality multi-character audiobook can take over 30 days and cost thousands to tens of thousands of yuan.
The first product line of Wanzhuang Yousheng, "Track One: Enhancing Quality," was created to solve the problem of "doing better."
It provides a full-cycle AI creation tool for professional studios and voice acting teams. Modules like intelligent script creation, intelligent alignment, and intelligent listening free people from tedious, repetitive labor, allowing them to focus on artistic creation. The core logic remains deep human-machine collaboration.
The newly opened fully automatic AI multi-voice audiobook workstation belongs to "Track Two: Capacity Leap."
Its goal is entirely different: it is not for meticulous creators, but for B-end clients such as online literature platforms and publishing institutions that hold a large amount of copyright and need to convert inventory quickly. Its logic is more straightforward — use a fully automated solution to solve the scalability challenge of massive, long-tail, ROI-sensitive content.
After users upload manuscripts, the system functions like a super digital factory: automatically splitting chapters, intelligently analyzing characters and matching the best AI voice, generating scripts with emotional annotations and precise pronunciation, using nearly a thousand optimized voice tones for multi-character narration, and finally synthesizing the final product with sound effects. After generation, the work will be transferred to a standard editing interface, where users can use newly launched features like "custom pronunciation" and "bulk annotation" to make local refinements, achieving a flexible production model of "mass-producing rough drafts first, then making local adjustments."
7 Yuan for 10,000 Characters: Clearing a Disruptive Account
Wanzhuang Yousheng has set a highly competitive price: 7.9 yuan per 10,000 characters for AI voice acting, 0.2 yuan per 10,000 characters for intelligent script creation, and as low as 6.58 yuan per 10,000 characters for bulk purchases. This means the total cost for producing a 10,000-character multi-voice audiobook is only 7–8 yuan.
By comparison, traditional methods require 5,000 to 50,000 yuan for one audiobook; other AI tools on the market generally range between 10–15 yuan per 10,000 characters. For copyright platforms holding thousands of mid-length IP works, the traditional method would require investment of tens of millions or even hundreds of millions of yuan for complete audiobook conversion. Even if only 10% were converted, it would be a heavy burden. However, according to Wanzhuang Yousheng's pricing, the audiobook conversion cost for a 500,000-character novel would only be 350–400 yuan. A medium-sized publishing house could turn an entire category of inventory into audio assets for just a few ten-thousands of yuan.
Li Ji, founder of Yuemei Culture, a leading copyright holder in the field of Xiangguan Xiaoshuo, once commented when Wanzhuang Yousheng began its public test: "The AI voice acting of Wanzhuang Yousheng has already approached the level of real human performance in emotional delivery. Combined with intelligent script creation and automatic alignment, it can stably produce audiobooks of B+ level or higher."
This quality benchmark was transferred to the fully automatic production line, making "acceptable quality per production line" a prerequisite. The remaining core issue then becomes purely an economic calculation.
Focusing on B-End: Strategic Implications of the Dual-Track Engine
This fully automatic workstation was initially used internally to demonstrate "technical supremacy" to rights holders and investors. Now, it has been commercialized, sending a clear signal.
This marks the expansion of Wanzhuang Yousheng's reach from "serving creators" to "serving rights holders"—the latter being the core client group with significant payment capability and urgent demand. The platform's unique "dual-track production engine" is not merely a collection of features, but based on a deep industry insight: the audiobook market simultaneously faces two seemingly contradictory needs—premium quality and mass production. The real bottleneck lies in the lack of a standardized production system that can dynamically adapt to different content values and production demands.
· For individuals or small teams, it serves as an efficient "content validator," allowing novels to be quickly transformed into shareable demos without the need for a professional team, enabling low-cost market testing.
· For online literature platforms or copyright institutions, it directly addresses the dilemma of "tasteless but hard to give up" mid-length IP works, awakening dormant inventory at a minimal marginal cost and quickly filling content gaps.
· For professional audio studios, it opens up a new "human-machine hybrid" model. Complex emotional main characters can be left to real voice actors, while narrations and a large number of NPC roles can be handled by high-quality AI, ensuring auditory experience while significantly expanding production capacity.
The engineering barriers behind it, such as a million-word pronunciation error correction dictionary, character consistency algorithms, and intelligent redrawing technology (which can reduce computing power consumption by 90% for local modifications), are the result of deeply integrating audiobook industry production experience. This is not only a technological victory but also a clear definition of the industrialization path of audio production.
Having successfully tested the internal production line, the team dares to deliver to external clients. Their reach has expanded from "serving creators" to "serving rights holders"—the true customers with bulk payment capabilities.
For platforms like Yewen and Qimao, or traditional publishing groups, the real concern is not "whether we can create a high-quality audiobook," but "with thousands of works in our library, how can we cost-effectively and on a large scale convert them into audio content." This fully automatic workstation offers them a production line they can activate on demand. It is not intended for individual creators.
Copyright institutions have long decision-making cycles, and it takes time from initial contact to volume production. Traditional publishers are used to outsourcing models, and switching to AI automation requires internal process and quality control changes. However, the production cost of audiobooks is rapidly decreasing. When the cost of producing a 10,000-character audiobook drops from thousands of yuan to just a few yuan, the supply logic of the industry will inevitably change.
About Wanzhuang Yousheng
Wanzhuang Yousheng (Audimind) was founded in 2024 by the former core management team of Lazy Listen. It is an AI-powered one-stop platform for audiobook content creation. Through its dual-track production engine, the platform provides comprehensive solutions covering the entire audiobook creation process for professional creators and copyright institutions.
Website: https://www.audimind.com/
