Objective The purpose of this investigation was to compare the efficacy of an automatic speech recognition (ASR) system and traditional methods for evaluating speech intelligibility in children with hearing loss. Design Observational Research Outcome Setting Department of Otolaryngology-Head and Neck Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, China Participants Ninety- four children aged 30 to 180 months with hearing loss who used hearing aids or cochlear implants prior to the examination and children with hearing loss who had congenital microtia. Main outcome measures Speech intelligibility evaluation，include the speech intelligibility score of traditional evaluation and automatic speech recognition system. Results For the speech intelligibility evaluation in the 94 children, the ASR system yielded speech intelligibility scores ranging from 0 to 100 (mean 64.3±31.1), and the traditional method yielded speech intelligibility scores ranging from 0 to 100 (mean 81.5±24.4. The correlation between the results of the traditional method and those of the ASR system was 0.895 (p <0.001). Conclusion Speech intelligibility evaluation for children with hearing loss in the clinic can be performed using the ASR system. This method can increase the precision, objectivity, and efficacy of speech intelligibility assessments.