TL;DR:
- NetEase Youdao has released “EmotiVoice,” an open-source text-to-speech (TTS) engine available on GitHub.
- EmotiVoice boasts over 2,000 timbres and supports both Chinese and English languages.
- Its standout feature is emotional synthesis, allowing for a wide range of emotions in generated speech.
- User-friendly online and scripting interfaces enable bulk result generation.
- EmotiVoice requires an NVidia graphics processing unit for testing.
- It shifts from gender-based to emotion-focused user prompts.
- This innovation enhances the text-to-speech technology landscape.
Main AI News:
In a significant move, NetEase Youdao has officially launched “Yi Mo Sheng,” an open-source text-to-speech (TTS) engine, now available on GitHub. This powerful tool, equipped with web and script interfaces, empowers users to generate results in bulk, catering to the demands of applications that require nuanced emotional synthesis.
Youdao, the mastermind behind this groundbreaking text-to-speech engine, has curated an extensive collection of more than 2,000 timbres, ensuring seamless support for both Chinese and English languages. What sets this engine apart is its unique emotion synthesis feature, enabling the creation of speech imbued with a diverse spectrum of emotions, including joy, excitement, sadness, and anger, accompanied by an array of expressive vocalizations.
In the realm of open-source text-to-speech engines, EmotiVoice reigns supreme. With a repertoire of over 2,000 distinctive voices and the ability to converse fluently in both English and Chinese, it offers an unparalleled experience. Its standout feature, emotional synthesis, empowers users to infuse their speech with a wide range of emotions, from elation and enthusiasm to sorrow and indignation.
EmotiVoice offers a user-friendly online interface for effortless operation, while the scripting interface facilitates the generation of results at scale. To put EmotiVoice to the test, users will require a computer equipped with an NVidia graphics processing unit. Installation of the NVidia container toolkit on Linux or Windows WSL2 is recommended if not already in place.
In the current landscape, user prompts play a pivotal role in shaping emotional responses and actions. EmotiVoice shifts the focus from gender to emphasize tone, tempo, intensity, and passion. Additionally, the engine allows for the incorporation of a style/timbre controller, mirroring the convenience of the original closed-source design. With EmotiVoice, the future of text-to-speech technology looks more expressive and emotionally engaging than ever before.
Conclusion:
The launch of EmotiVoice by NetEase Youdao represents a significant advancement in the text-to-speech market. With its open-source approach, extensive timbre collection, and emotional synthesis capabilities, EmotiVoice is set to redefine how users engage with text-to-speech technology. Its emphasis on emotion-based prompts aligns with the growing demand for more expressive and emotionally engaging communication tools, making it a game-changer in the market.
