I have experimented a lot for human like text to Speech Api’s. I have already used Google Wavenet & Standard, Amazon & Microsoft neural voice, Yandex etc to create a more human speech voice using text to speech api’s for Voice bot. I have also experimented a lot using SSML but so far none of them sound human.
Does anyone know any services, no matter what is the cost that can help me get real time Human Sounding AI voices to be integrated with RASA. I m also in contact with Google and Microsoft for Custom Voice solution. In case I have missed something brilliant available please let me know.
I have seen these already and have actually been able to create better bots than these using ssml. More synthetic voices. But my most focus is on getting a custom AI voice that doesnt sound inhuman (monotonous).
Hey, can you somehow help in connecting my rasa bot with Microsoft speech container? Probably give an overview of how you achieved it, would be really helpful.
