AliTongyi promotes a new generation of voice model Fun-ASR

On August 22nd, it was reported that Alibaba's communications platform AliTalk has launched a new generation end-to-end speech recognition model called Fun-ASR. This model enhances context awareness and high-precision speech transcription capabilities, improving the accuracy of speech recognition by over 15% in various industries such as home decoration and insurance. Currently, Fun-ASR has been applied in scenarios such as conference subtitles, simultaneous interpretation, intelligent minutes, and virtual assistants. In the future, this model will be further launched on Alibaba Cloud Hundred-level.