Measuring the Effectiveness of Voice Conversion on Speaker Identification and Automatic Speech Recognition Systems

Voice Conversion

Categories: Generative Adversarial Network

Tags: 2019 | Cory Stephenson | Gokce Keskin | Oguz H. Elibol | Tyler Lee

This paper evaluates the effectiveness of a Cycle-GAN based voice converter (VC) on four speaker identification (SID) systems and an automated speech recognition (ASR) system for various purposes. Audio samples converted by the VC model are classified by the SID systems as the intended target at up to 46% top-1 accuracy among more than 250 speakers. This encouraging result in imitating the target styles led us to investigate if converted (synthetic) samples can be used to improve ASR training. Unfortunately, adding synthetic data to the ASR training set only marginally improves word and charac...

Many-to-Many Voice Conversion with Out-of-Dataset Speaker Support

Voice Conversion

Categories: Adversarial Training | Convolutional Neural Network | Deep Learning | Fourier Transform | Generative Adversarial Network | Mean Opinion Score | Mel-spectrogram

Tags: 2019 | Cory Stephenson | Gokce Keskin | Oguz H. Elibol | Tyler Lee

We present a Cycle-GAN based many-to-many voice conversion method that can convert between speakers that are not in the training set. This property is enabled through speaker embeddings generated by a neural network that is jointly trained with the Cycle-GAN. In contrast to prior work in this domain, our method enables conversion between an out-of-dataset speaker and a target speaker in either direction and does not require re-training. Out-of-dataset speaker conversion quality is evaluated using an independently trained speaker identification model, and shows good style conversion characteris...