Segmentation fault
/zywx/workspace/hotel_rec/.venv/lib/python3.10/site-packages/torch/nn/utils/weight_norm.py:143: FutureWarning: torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.
WeightNorm.apply(module, name, dim)
/zywx/workspace/hotel_rec/.venv/lib/python3.10/site-packages/torch/nn/modules/rnn.py:123: UserWarning: dropout option adds dropout after all but last recurrent layer, so non-zero dropout expects num_layers greater than 1, but got dropout=0.2 and num_layers=1
warnings.warn(
Loaded voice: af
Segmentation fault (core dumped)
Those are Warnings, they are not Error messages, so there is not much context for that segfault other than it is happening after the voicepack loads (so build_model should be fine).
Assuming this is on your local machine:
- First, get it working on Colab. It should be copy-paste-run.
- Check
pip show torch transformersandespeak-ng --versionon your local install and compare these versions to Colab. Minor version deltas are probably fine, major ones are not. - If you are familiar with code, some basic print statements in
kokoro.pyundergenerateshould help you narrow down where the segfault is happening. It's 2025 and you can use LLMs to help you debug, whether you know how to code or not.
This seems like a case of debugging 101 so I will close this issue, hop on Discord if you need more help. Will reopen if the segfault is happening in Colab.