Alibaba's New Voice Model Takes Top Spot
The new Qwen Audio 3.0 TTS Plus model from Alibaba is now the top performer on the Speech Arena leaderboard. It works in 16 different languages and gives users precise control over the tone of the voice. You can change how it sounds by using simple tags like angry or by describing the style in plain English.
While the quality is high, the model is not very fast. It generates speech at a rate of 16 characters per second. This is much slower than other popular options like Sonic 3.5 and Simba 3.2.
This technology represents a big step forward for Alibaba in the voice space. Users will have to decide if the expressive output is worth the wait compared to faster competitors.
Comments (0)
No comments yet. Be the first!
More AI news
NewsAlibaba's New AI Can Finally Write Tiny Text
Alibaba just released an image generator that creates readable text and complex charts in one go.
NewsMicrosoft and Mistral Team Up for European AI
Microsoft is investing billions to help Mistral build new AI technology throughout Europe.
NewsWhy AI Feels Like a Slot Machine
Using AI tools can trap you in a cycle of endless tweaks that kills your productivity.