Shulou(Shulou.com)11/24 Report--

On November 10, NetEase youdao officially launched the "easy Magic Sound" open source speech synthesis (TTS) engine, which can be downloaded and used by all users in the open source community GitHub for free. Through its web interface and script interface for batch generation of results, it is easy to achieve emotional synthesis and application of timbre.

It is reported that easy Magic Sound is a youdao self-developed TTS engine, which currently supports both Chinese and English, including more than 2000 different timbre, more characteristic emotion synthesis function, and supports the synthesis of voice that includes a wide range of emotions such as happiness, excitement, sadness and anger.

(GitHub open source interface)

In the memories of the past, there are always some special voices, such as: the voice of an idol is inspiring, and the voice of our mother reminds us of our childhood in a second. Sound, as a kind of language dimension, always contains abundant emotional expression of human beings. The emotional synthetic voice is an AI function that can add color to the application and content. Youdao "easy Magic Voice" provides some solutions for developers and content creators-simply by adding emotional description hints to the text, they can freely synthesize emotional voice that meets their needs, which is more natural and lifelike than traditional TTS.

As the voice ability of modern AI technologies such as GAN becomes more and more mature, the threshold for implementing a high-quality TTS system is getting lower and lower. But even so, it is difficult to find high-quality and modern TTS modules in both Chinese and English, and it is still troublesome to add high-fidelity and highly controllable speech to their own applications and content, especially in both Chinese and English.

"at present, the project is still in its early stages, and now open source the project, hoping to help developers and content creators in need, and continue to expand the scope of application of high-quality TTS, so that products and applications can be better landed. We also look forward to providing us with more feedback and suggestions after the trial." NetEase youdao CEO Zhou Feng said.

Youdao has worked hard in the field of TTS for many years, and has always been scene-oriented, constantly promoting the landing of technology, bringing many efficient and convenient applications and products for users. For example, the first star voice function in the field of education is launched, in which the voices of stars such as Wang Yuan, Ouyang Nana and Ma Boqian are built into NetEase's youdao dictionary to accompany users to learn English together. Voice customization and voice reproduction functions are provided, and personalized voice customization can be completed in only 5 minutes. The recently launched private oral teaching of Hi Echo virtual human, with the help of youdao "Zi Yue" education big model, pronunciation and virtual human technology, helps users to practice oral English easily anytime, anywhere.

Since 2008, NetEase youdao has begun to lay out AI, has been committed to innovation and application based on Transformer model for many years, and has core technologies in neural network translation, computer vision, high-performance computing, intelligent voice AI technology, etc., which has laid a solid technical foundation for the actual landing of the application.

In addition, through youdao Zhiyun's official website, users can experience various AI technologies such as text and image translation, text and image recognition, composition correction, which are already open to developers through API.

Youdao Zhiyun AI open platform is an one-stop artificial intelligence service provider under NetEase youdao. It provides natural language translation, text recognition, OCR, voice recognition and other services as well as industry solutions for developers, enterprises and government agencies, and is committed to providing secure, reliable and efficient cloud services.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.