-
Since the coqui docs recommend the use of `deepspeed` to speed up their XTTS model I wanted to give this a try.
To make it work I did the following:
- I had to rebuild pytorch with `USE_NCCL=1` be…
hslr4 updated
3 weeks ago
-
Hi! First, i want to thank you for making this repo and sharing your experiences.
I am trying to train XTTS model for Persian language. I read the topics in this repo https://github.com/coqui-ai/TTS/…
-
Current script directory: E:\waifu\Waifu-texto-ollama-xtts\
E:\waifu\Waifu-texto-ollama-xtts\xtts-venv\lib\site-packages\pydub\utils.py:170: RuntimeWarning: Couldn't find ffmpeg or avconv - defaultin…
-
### Describe the bug
Hello, training XTTSv2 leads to weird training lags - training gets stuck with no errors
with using DDP
x6 RTX a6000 and 512GB RAM
Here is monitoring GPU load graph. Purpl…
-
测试音频,请参考附件,音频25s,能正常播放,但识别时报错
File "/home/roots/data/SenseVoice/webui.py", line 167, in model_inference
text = model.generate(input=input_wav,
File "/home/roots/anaconda3/envs/coqui-x…
-
**Description:**
When running the start.sh and install.sh scripts, I encountered an AttributeError in xtts_demo.py. The error occurs due to an incorrect import of GradScaler from torch.amp.
**Step…
-
### Describe the bug
When using the [google colab](https://colab.research.google.com/drive/1GiI4_X724M8q2W-zZ-jXo7cWTV7RfaH-?usp=sharing) for XTTS_v2 that is linked in the [official tutorial](https…
-
After download the [model.bin] I get following error:
> tts_models/multilingual/multi-dataset/xtts_v2 is already downloaded.
> Using model: xtts
model.bin: 100%|█████████████████████████████████…
-
After adding a 5 minute wav / mp3 file and clicking on Step 1 - Create Dataset. I get this error:
The processing was interrupted due an error !! Please check the console to verify the full error me…
-
Traceback (most recent call last):
File "C:\MR\Finetune\xtts-finetune-webui\venv\lib\site-packages\gradio\queueing.py", line 459, in call_prediction
output = await route_utils.call_process_api…