I have tested its speed on CPU and compared it with Piper, kokoro, and matcha. See https://github.com/KittenML/KittenTTS/issues/40
I am one of the authors of sherpa-onnx. Can you describe why you feel it is complex? If you use Python, all you need is to run pip install sherpa-onnx, and then download a model and use the example code from the folder…
I have tested its speed on CPU and compared it with Piper, kokoro, and matcha. See https://github.com/KittenML/KittenTTS/issues/40
I am one of the authors of sherpa-onnx. Can you describe why you feel it is complex? If you use Python, all you need is to run pip install sherpa-onnx, and then download a model and use the example code from the folder…