West Lake invested by Jinke TomcatXinchenIn August this year, Xinchen was launched Lingo Voice big model, the first end-to-end speech model in China, was launched today (August 24)Closed betareserve.
In the announcement released on August 21, the official introduction stated that compared with traditional TTS, the end-to-end speech big model is a more comprehensive technology. It not only can recognize speech, but also integrates natural language processing, intent recognition, dialogue management, and speech synthesis. It realizes the complete interactive process from speech input to speech feedback, greatly enriching the depth and breadth of human-computer interaction.
Xinchen Lingo voice model is the first model in China that has the same voice capabilities as GPT-4o, the technical capabilities have the following three notable characteristics:
- Native speech understanding:As an end-to-end model, Lingo can not only recognize text information in speech, but also accurately capture other important features such as emotion, tone, pitch, and even ambient sound, helping the model to understand the speech content more comprehensively, thereby providing a more natural and vivid interactive experience.
- Multiple voice styles:Lingo can adaptively adjust the speed, pitch, and noise intensity of speech according to the context and user instructions, and can generate voice responses in a variety of styles such as conversation, singing, and crosstalk, effectively improving the flexibility and adaptability of the model in different application scenarios.
- Voice Mode Super Compression:Lingo uses a voice codec with a compression rate of hundreds of times, which can compress voice to an extremely short length, significantly reducing computing and storage costs while helping the model generate high-quality voice content.