search

torchsynth / torchsynth

A GPU-optional modular synthesizer in pytorch, 16200x faster than realtime, for audio ML researchers.
https://torchsynth.rtfd.io/
Apache License 2.0
321 stars 11 forks source link

synth sound talking #372

Open taalua opened 2 years ago

taalua commented 2 years ago

is there any example to make the synth sound talking?

jorshi commented 2 years ago

@taalua thanks for your question! I'm not quite sure what you are referring to regarding the synth sound talking? Do you mean speech synthesis?

torchsynth is currently focused on digital audio synthesis for machine learning. The architecture and dsp modules are inspired by musical synthesizers. -- we have examples of sounds from the default voice here: https://soundcloud.com/user-357924775/synth1k1 -- or an example of how to start creating your own synths using the available modules here: https://torchsynth.readthedocs.io/en/latest/modular-design/new-synths.html

taalua commented 2 years ago

@jorshi thank you for your answer. I am wondering if there is a way to mix speech and the synthesizers together to create a robotic voice or chipmunks for example. Do you have such examples?

turian commented 1 year ago

@taalua Are you interested in contributing?

You might also consider looking pink trombone which we would like to integrate into torchsynth.

Or, if you want something ready made, there are services like fake you