Open nukes opened 1 year ago
Hi, This is a nice work! Did you test the inference speed? how many tokens/second the model can generate on your GPU? i want to see if this model can generate audio much faster than realtime.
Hi, This is a nice work! Did you test the inference speed? how many tokens/second the model can generate on your GPU? i want to see if this model can generate audio much faster than realtime.