SYSTEM AND METHOD FOR VOICE UNIDENTIFIABLE MORPHING
A system and a method are disclosed for a machine learned audio morpher that is trained such that the voice characteristics of a user spoken phrase are replaced with those of a target speaker, which removes and/or reduces the user identifiable information for the spoken phrase. Training can be performed by a user and a target speaker speaking the same or similar phrases and training the audio morpher to minimize the differences between the target speaker phrase and a morphed user phrase.
1 . A method for training an audio morpher, the method comprising:
receiving a target speaker spoken phrase;
receiving a user spoken phrase;
computing a speaker representation of the target speaker;
morphing, using the audio morpher, the user spoken phrase based on the speaker representation;
creating a spectrogram of the target speaker spoken phrase;
creating a spectrogram of the morphed user spoken phrase; and
training the audio morpher with an objective function of minimizing the differences between the spectrogram of the morphed user spoken phrase with the spectrogram of the target speaker spoken phrase.
2 . The method of claim 1 , wherein the target speaker spoken phrase and the user spoken phrase use the same words.
3 . The method of claim 2 , wherein training of the audio morpher includes aligning the target speaker spoken phrase and the user spoken phrase.
4 . The method of claim 1 , wherein the words in the target speaker spoken phrase and the words in the user spoken phrase are similar. 6 . The method of claim 5 , wherein the target speaker spoken phrase and the user spoken phrase are different by less than a predefined number of phonemes.
5 . The method of claim 4 , wherein the percent difference between the number of phonemes in the target speaker spoken phrase and the user spoken phrase is below a predefined percentage.
6 . The method of claim 4 , wherein training of the audio morpher includes aligning the target speaker spoken phrase and the user spoken phrase.
7 . A system comprising a processor and memory, wherein the memory stores code that is executed by the processor to cause the system to morph audio, with an audio morpher, from a user spoken phrase to replace voice characteristics of a user spoken phrase with a target speaker's voice characteristics, wherein the audio morpher is trained using an objective function of minimizing the differences between the spectrogram of the morphed user spoken phrase with the spectrogram of the target speaker.
8 . The system of claim 7 , wherein the target speaker voice characteristics are selected at random from an array of target speaker representations.
9 . A method comprising:
receiving audio of a speaker's spoken words, which includes the speaker's personally identifiable information;
determining a pitch feature vector for the audio;
determining a phoneme representation for the audio;
receiving a fingerprint for a target speaker; and
generating morphed audio based on the pitch feature vector, the phoneme representation, and the fingerprint,
wherein the morphed audio protects the speaker's personally identifiable information associated with the speaker's voice.
10 . The method of claim 9 , wherein the target speaker is randomly selected from a plurality of target speakers.
11 . The method of claim 9 , wherein the fingerprint for the target speaker is determined using an ECAPA.
12 . The method of claim 9 , wherein the morphed audio is generated by concatenating the fingerprint, the pitch feature vector, and the morphed audio.