English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Text-to-speech (TTS) systems are being built using end-to-end deep learningapproaches. However, these systems require huge amounts of training data. Wepresent our approach to built production quality TTS and perform speakeradaptation in extremely low resource settings. We propose a transfer learningapproach using high-resource language data and synthetically generated data. Wetransfer the learnings from the out-domain high-resource English language.Further, we make use of out-of-the-box single-speaker TTS in the targetlanguage to generate in-domain synthetic data. We employ a three-step approachto train a high-quality single-speaker TTS system in a low-resource Indianlanguage Hindi. We use a Tacotron2 like setup with a spectrogram predictionnetwork and a waveglow vocoder. The Tacotron2 acoustic model is trained onEnglish data, followed by synthetic Hindi data from the existing TTS system.Finally, the decoder of this model is fine-tuned on only 3 hours of targetHindi speaker data to enable rapid speaker adaptation. We show the importanceof this dual pre-training and decoder-only fine-tuning using subjective MOSevaluation. Using transfer learning from high-resource language and syntheticcorpus we present a low-cost solution to train a custom TTS model.

Rapid Speaker Adaptation in Low Resource Text to Speech Systems using Synthetic Data and Transfer learning