This study explores the acceptability of a humanoid robot in the homes of deaf children with cochlear implants. Traditional audiologic rehabilitation, usually conducted at hospitals or clinics, significantly improves communication but geographical distance and parental availability pose challenges for families. Introducing home training with a humanoid robot could provide greater equity in care distribution, aid in the child’s progress, and offer an ecological approach. Here, ten families (11 cochlear-implanted (CI) children aged 7–12 years and their parents) participated in a study evaluating the acceptability of the humanoid robot (called Pepper) in their homes for one month. We identified two distinct clusters among the 11 CI children. The first cluster ("younger CI Children") demonstrated better and higher scores for acceptability, usability, usefulness, emotion, and fun, and used the robot more than children in the second cluster ("older CI children"). We found a decrease in both duration of use and number of applications launched by CI children each week when interacting with the robot. Parents showed similar results to their children, with parents of younger CI children demonstrating greater acceptance of the robot. Various technical issues (transcription errors, robot non-responsiveness, and comprehension difficulties) may have affected the robot’s acceptability among both CI children and their parents. Both CI children and parents expected humanoid robots to have more advanced capabilities, including high social intelligence, advanced artificial intelligence, voice assistant features, and Natural Language Processing integration.