The 7th Sound Expo kicks off and releases a series of AI applications
Author：Eve Cole
Update Time：2024-11-16 18:24:02
On October 24, the 7th World Sound Expo and the 2024 iFlytek Global 1024 Developer Festival opened in Hefei. iFlytek Chairman Liu Qingfeng announced the iFlytek Spark large model application report card, and released iFlytek Spark 4.0 Turbo and Related applications and products that empower people’s livelihood. On the same day, the domestic ultra-large-scale intelligent computing platform "Feixing 2" jointly built by iFlytek, Huawei, and Hefei Big Data Asset Operation Co., Ltd. was officially launched. Liu Qingfeng introduced that the three-party joint team has overcome many "difficult diseases" in the past year and solved more than 500 basic software and hardware problems and model adaptation problems. In the future, "Feixing 2" will bring new models and new algorithms. Continuous adaptation and scale development of intelligent computing clusters. At the scene, the super-anthropomorphic digital human created by iFlytek made its debut, realizing multi-modal interaction of voice, video, image and text, and supporting users to create their own personalized digital human with simple editing and definition in the background. You can quickly generate your own cartoon image. It is worth mentioning that in the field of speech recognition, iFlytek’s far-field high-noise scene speech recognition technology has further expanded its advantages. In terms of multi-lingual capabilities, for the first time, it has achieved full coverage of more than 200 dialects in prefecture-level cities across the country; in terms of multi-lingual capabilities, it has released the Spark multi-language large model for the first time, which in addition to Chinese and English, can support Russian, Japanese, Arabic, French, etc. 8 language. At the scene, Huawei and iFlytek jointly launched an innovative technology - the sound repair function, which uses powerful real-time voice processing capabilities. When users pronounce words, their speech is analyzed in real time and repaired and optimized to improve the intelligibility and clarity of pronunciation to help people with speech impairments achieve smoother communication.
