nodepositbonuscryptocasino| China Telecom releases its first large voice model that supports mixed speaking in 30 dialects

News summary

[China Telecom releases its first large voice model that supports mixed speech in 30 dialects] Securities Times e Company NewsnodepositbonuscryptocasinoRecently, China Telecom Artificial Intelligence Research Institute released the industry's first large speech recognition model that supports the free mixing of 30 dialects-Xingchen Super Multi-Dialect Speech Recognition Model, breaking the single.nodepositbonuscryptocasino...

Newsletter text

[China Telecom releases its first large speech model that supports the mixed spoken of 30 dialects] Securities Times e Company News that recently, China Telecom's Artificial Intelligence Research Institute released the industry's first large speech recognition model that supports the free mixing of 30 dialects-Xingchen Super The multi-dialect speech recognition model breaks the dilemma that a single model can only recognize a specific single dialect, and can simultaneously recognize and understand more than 30 dialects such as Cantonese, Shanghai, Sichuan, and Wenzhou dialects. It is the large speech recognition model that supports the largest number of dialects in China. It is reported that the large-scale model research and development team pioneered the "distillation + expansion" joint training algorithm to solve the problem of pre-training collapse under ultra-large-scale multi-scene data sets and large-scale parameters, and achieve stable training of 1B parameter 80-layer models. Xingchen Speech Model is also the industry's first open source speech recognition model based on discrete speech representations. Through a new modeling paradigm of "from speech to token to text", it reduces the speech transmission bit rate during reasoning by dozens of times. At present, the Star Voice Model has been piloted in China Telecom Wanhao intelligent customer service in Fujian, Jiangxi, Guangxi, Beijing, Inner Mongolia and other places.