Double dipping is a common practice in voice banking and voice preservation for ALS patients. However, using the same recorded speech material for both training a speech synthesis model and for direct use as speech output can result in several problems. The most significant issue is that it reduces the quantity and quality of the recorded speech material available for the patient's own use, as the same recordings are being used for multiple purposes. This can limit the patient's ability to express themselves effectively and can also impact the effectiveness of the speech synthesis model.
On the other hand, having two separate recordings sets - one specifically for training the speech synthesis model and another for direct use as speech output - can offer several advantages. Firstly, it ensures that the recorded speech material is being used in the most effective and efficient manner possible, maximizing the patient's speech resources. Secondly, having separate recordings for direct use can make the speech output more personal, as it allows the patient to choose the recordings that best represent their unique speech patterns and mannerisms. This can help to maintain the patient's identity and sense of self, even as their physical abilities change.
Today, when the required number of sentences to be recorded is rather small, and the time investment is usually much less than an hour, we recommend to first create a voice using the read out script, the “Voice Banking” process, and only then, add personal messages.
The benefit of starting with our script is that it is “the fast track” to preserving your voice. Not only that it’s fast and easy, it doesn’t require you to THINK about the messages to read, thus also loading the emotional load of creating a voice.
Double Dipping using The Voice Keeper:
Although not recommended as best practice, Double Dipping can be achieved in these ways:
Importing existing Message Banking set of audio files, as can be read here [link to message banking conversion page]
Creating personal messages / recording Message Banking DIRECTLY on The Voice Keeper platform, as can be read here: [link to b]