Govender, Avashna2022-10-032022-10-032022-05Govender, A. 2022. Multi-MelGAN voice conversion for the creation of under-resourced child speech synthesis. http://hdl.handle.net/10204/12496 .http://hdl.handle.net/10204/12496Voice conversion (VC) is an important technique for the development of text-to-speech voices in the use case of lacking speech resources. VC can convert an audio signal from a source speaker to a specific target speaker whilst maintaining the linguistic information. The benefit of VC is that you only require a small amount of target data which therefore makes it possible to build high quality text-to-speech voices using only a limited amount of speech data. In this work, we implement VC using a Melspectrogram Generatative Adversarial Network called MelGAN-VC. This technique does not require parallel data and has been proven successful on as little as 1 hour of target speech data. The aim of this work was to build child voices by modifying the original one-to-one MelGAN-VC model to a many-to-many model and determine if there is any gain in using such a model. We found that using a many-to-many model performs better than the baseline one-to-one model in terms of speaker similarity and the naturalness of the output speech when using only 24 minutes of speech data.FulltextenVoice conversionText-to-speech voicesMelGAN-VCMelspectrogram Generatative Adversarial NetworkSpeech dataConference PresentationGovender, A. (2022). Multi-MelGAN voice conversion for the creation of under-resourced child speech synthesis. http://hdl.handle.net/10204/12496Govender, Avashna. "Multi-MelGAN voice conversion for the creation of under-resourced child speech synthesis." <i>IST-Africa 2022 Conference Proceedings, Virtual, South Africa, 16-20 May 2022</i> (2022): http://hdl.handle.net/10204/12496Govender A, Multi-MelGAN voice conversion for the creation of under-resourced child speech synthesis; 2022. http://hdl.handle.net/10204/12496 .TY - Conference Presentation AU - Govender, Avashna AB - Voice conversion (VC) is an important technique for the development of text-to-speech voices in the use case of lacking speech resources. VC can convert an audio signal from a source speaker to a specific target speaker whilst maintaining the linguistic information. The benefit of VC is that you only require a small amount of target data which therefore makes it possible to build high quality text-to-speech voices using only a limited amount of speech data. In this work, we implement VC using a Melspectrogram Generatative Adversarial Network called MelGAN-VC. This technique does not require parallel data and has been proven successful on as little as 1 hour of target speech data. The aim of this work was to build child voices by modifying the original one-to-one MelGAN-VC model to a many-to-many model and determine if there is any gain in using such a model. We found that using a many-to-many model performs better than the baseline one-to-one model in terms of speaker similarity and the naturalness of the output speech when using only 24 minutes of speech data. DA - 2022-05 DB - ResearchSpace DP - CSIR J1 - IST-Africa 2022 Conference Proceedings, Virtual, South Africa, 16-20 May 2022 KW - Voice conversion KW - Text-to-speech voices KW - MelGAN-VC KW - Melspectrogram Generatative Adversarial Network KW - Speech data LK - https://researchspace.csir.co.za PY - 2022 T1 - Multi-MelGAN voice conversion for the creation of under-resourced child speech synthesis TI - Multi-MelGAN voice conversion for the creation of under-resourced child speech synthesis UR - http://hdl.handle.net/10204/12496 ER -25900