Asynchronous Voice Anonymization by Learning From Speaker-Adversarial Speech

This letter focuses on asynchronous voice anonymization, wherein machine-discernible speaker attributes in a speech utterance are obscured while human perception is preserved. We propose to transfer the voice-protection capability of speaker-adversarial speech to speaker embedding, thereby facilitating the modification of speaker embedding extracted from original speech to generate anonymized speech. Experiments conducted on the LibriSpeech dataset demonstrated that compared to the speaker-adversarial utterances, the generated anonymized speech demonstrates improved transferability and voice-protection capability. Furthermore, the proposed method enhances the human perception preservation capability of anonymized speech within the generative asynchronous voice anonymization framework.