Recently, Qifu Technology was invited to attend the top international conference on speech communication and signal processing "INTERSPEECH 2024" held in Greece, and delivered a keynote speech entitled "Qifusion Net: Layer adapted Stream/On stream Model for End to End Multi Accent Speech Recognition", fully demonstrating its achievements in the field of speech recognition technology.
As one of the most prestigious annual conferences in the global speech science community, INTERSPEECH brings together top scholars, researchers, and industry leaders from around the world to discuss the latest developments, challenges, and future trends in speech technology.
In the speech, QiFree Technology introduced its new generation QiFree speech recognition system, which can support over 20 dialects simultaneously. In the comparison of the authoritative test set KeSpeech in the field of Chinese accent and dialect speech recognition, Qifu Technology has achieved a significant improvement in dialect accent classification accuracy, reaching 79.10%, thanks to its accumulation in automatic speech recognition. Meanwhile, in terms of the key indicator CER (Character Error Rate) for measuring recognition error rate, Qifu Technology has demonstrated its efficiency and accuracy in the field of Chinese dialect recognition.