This distribution includes synthetic voices of 13 Indian Languages. These voices are a results of the joint effort by the consortium partners. Each of the language has at least one of the following type of voices: 1. HTS (pruned) 2. HTS (unpruned) 3. USS (unpruned) 4. USS (pruned). For HTS voice models, atmost 3 different flavors are available for a language. (a) HTS-2.2 : Voice model built using HTS 2.2 and Hybrid Segmentation. (b) HTS-2.3 : Voice model built using HTS 2.3 and Hybrid Segmentation (c) HTS-2.3-DNN (5hr) : Voice model built using HTS 2.3 and DNN Segmentation with 5hrs of data for training (d) HTS-2.3-DNN (10hr) : Voice model built using HTS 2.3 and DNN Segmentation with 10hrs of data for training
Indic TTS : Enhancing text-to-speech (TTS) synthesis for Indian languages, optimizing quality and integrating compact TTS into disability aids and diverse applications.