• Subscribe
  • Ask me anything related to text to speech, voiceover, dubbing and subtitle industry

    Narasimha Suda
    8 replies

    Replies

    Oliver Han
    What do you think about my product: https://www.voxwaveai.com/
    Narasimha Suda
    @okhannaford This space is getting hot for sure! Video personalization is the best way forward to increase engagement and conversions however, finding a niche in the target segment and integration with the existing stack will bring them close to adoption.
    Harinderpreet singh
    How much it will cost you build Natural sounding TTS like wellsaidlabs? Is it possible to have mimic style while generating voiceover using tts (basically we can upload sample and it will mimic our voice style)? Almost all tts websites sound same (I can they are similar to each other) I see no advanced in companies compared to each other so how does future look like ? How difficult is to build voice clone model that can run locally on computer
    Nikhil Sharma
    @narasimha_suda what are some of the key factors that impact the quality of AI-generated speech?
    Narasimha Suda
    @imnikhill10 Primary factor is the quality of training data. If you have high-quality training data, the synthesis will be much better. Second, the model in which you train the model. More controllable parameters such as pitch, pace, emotion, tone, etc will dramatically enhance the quality.
    Nikhil Sharma
    @narasimha_suda In addition to controllable parameters such as pitch, pace, and emotion, what other factors can impact the quality of text to speech synthesis, and how can they be addressed?