lukaLLM
lukaLLM
The solution above worked for me thanks so much using the nvidia newest container nvcr.io/nvidia/pytorch:24.08-py3
Then I got the error ImportError: /usr/local/lib/python3.10/dist-packages/fused_layer_norm_cuda.cpython-310-x86_64-linux-gnu.so: undefined symbol: _ZN3c106detail14torchCheckFailEPKcS2_jRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE the solution is here https://github.com/facebookresearch/fairseq/issues/4246
Same windows11 Python 3.11.9 The conflict is caused by: piper-tts 1.2.0 depends on piper-phonemize~=1.1.0 piper-tts 1.1.0 depends on piper-phonemize~=1.0.0 To fix this you could try to: 1. loosen the range...
Thanks, you helped me to make it work on windows!
Hey I am not sure if this got finished or not. I am working with faster whisper a bit and could try to make it async as I need it...
I have same issue
@aixingxy I am getting the same issue It was working before on an old installation that I have on Conda it seems there was some update that made it happen...
@ylacombe Thanks I didn't know it existed tested Dutch with the default speaker and asked multiple people but the voice is like 5/10. But maybe we manage to collect some...
> This is a working snippet to continue audio in the same style as a speaker. In the snippet, we require: > > * `init_audio_file`: path to an audio file...
> @lukaLLM which branch r u on? You need to be in the aforementioned branch to be able to run this code. Okay my bad got some dependecy issue I...