Speech is the most common way of communication. People who have lost total or partially their ability to speak might benefit from the use of Alternative and Augmentative Communication (AAC) devices and the use of Text-to-Speech (TTS) technology. One problem that arouses is that the synthetic voices included in these devices might be impersonal and not accurate to the user terms of age, accent or even gender. Therefore, voice banking has become a good alternative to standard commercial voices. In our voice banking strategy, people with healthy voice (donors), or the user itself before losing his or her own voice, provide the recordings to obtain a new synthetic voice using adaptation techniques. In this way, a wide catalog of synthetic voices is provided to the potential user. However, because there is no control over the recording process, the final quality of the synthetic voice is very variable. In this paper, we propose a method to assess the result of the adaptation using objective measures. The results show that this strategy can be an alternative to subjective evaluation to select the best donated voices for the voice bank.