येन इदं महतोपायः (TTS) कृतः तस्मै श्री प्रतोषाय नमः॥
Let me start with my feedback that the output of this 'Sanskrit Shloka TTS' is really very natural and that it was much more than I had expected.
It would be useful for all of us to also know whether this can convert other types of Chanda's (छन्द ) and even gadya (गद्य ) into a speech of similar quality. Was it created by training on some AI model or some other technology/methodology was used for this? Can it also be improved to produce sound of a lady or a young boy?
What does Dr. Pratosh think is best way to produce a TTS for a new low resource language? Is making TTS much easier (compared to English, French etc) for most of Indian scripts which are 'phonetic' in nature?
-- अनुनाद