Bluebomber182
  • Joined on 2023-03-06
Bluebomber182 commented on issue mrq/ai-voice-cloning#454 2024-04-27 19:29:41 +00:00
StyleTTS2 - New Free Voice Cloning TTS

I found an improved styletts2 base model for those interested in fine tuning. https://huggingface.co/ShoukanLabs/Vokan

Bluebomber182 commented on issue mrq/ai-voice-cloning#454 2023-11-29 18:33:44 +00:00
StyleTTS2 - New Free Voice Cloning TTS

You can use the Inference_LibriTTS.ipynb file locally Just install these

pip install notebook jupyter Then open the Inference_LibriTTS.ipynb file jupyter notebook Inference_LibriTTS.ipynb

Bluebomber182 opened issue mrq/ai-voice-cloning#438 2023-11-01 20:33:28 +00:00
Request: Implement StyleTTS2 by yl4579?
Bluebomber182 opened issue mrq/vall-e#10 2023-11-01 04:18:43 +00:00
Request: Support for Descript Audio Codec
Bluebomber182 opened issue mrq/ai-voice-cloning#425 2023-10-23 18:30:49 +00:00
CUDA out of memory error after installing deepspeed
Bluebomber182 commented on issue mrq/ai-voice-cloning#424 2023-10-23 17:18:05 +00:00
RuntimeError: Error building extension 'transformer_inference'

I figured out the problem. First I activate the python enivroment. source ./venv/bin/activate Then I installed torch torchvision torchaudio with command. `pip3 install torch torchvision…

Bluebomber182 closed issue mrq/ai-voice-cloning#424 2023-10-23 17:18:05 +00:00
RuntimeError: Error building extension 'transformer_inference'
Bluebomber182 opened issue mrq/vall-e#9 2023-10-22 03:00:01 +00:00
AttributeError: 'AR_NAR' object has no attribute 'langs_emb'
Bluebomber182 opened issue mrq/ai-voice-cloning#424 2023-10-22 01:58:11 +00:00
RuntimeError: Error building extension 'transformer_inference'
Bluebomber182 closed issue mrq/ai-voice-cloning#395 2023-09-21 14:22:52 +00:00
Why does my AMD GPU eat up too much vram?
Bluebomber182 opened issue mrq/ai-voice-cloning#395 2023-09-20 06:59:58 +00:00
Why does my AMD GPU eat up too much vram?
Bluebomber182 closed issue mrq/vall-e#7 2023-08-30 18:07:28 +00:00
Training error: ValueError: num_samples should be a positive integer value, but got num_samples=0
Bluebomber182 opened issue mrq/vall-e#7 2023-08-30 14:14:19 +00:00
Training error: ValueError: num_samples should be a positive integer value, but got num_samples=0
Bluebomber182 commented on issue mrq/vall-e#6 2023-08-27 02:58:31 +00:00
Training error: RuntimeError: Could not infer dtype of NoneType

I created the HDF5 dataset. I still got the "RuntimeError: Could not infer dtype of NoneType" error message. I tried setting to "True" under use_hdf5: in config.yaml and I got this error…

Bluebomber182 opened issue mrq/vall-e#6 2023-08-27 00:49:08 +00:00
Training error: RuntimeError: Could not infer dtype of NoneType
Bluebomber182 closed issue mrq/ai-voice-cloning#324 2023-08-25 01:24:10 +00:00
Does bark feature fine tuning?
Bluebomber182 closed issue mrq/vall-e#5 2023-08-24 23:31:07 +00:00
Question about saving and quitting vall-e
Bluebomber182 commented on issue mrq/vall-e#5 2023-08-24 23:31:06 +00:00
Question about saving and quitting vall-e

Thank you! That did it! Typing in quit and pressing enter now works too.