auracaster/piper
3
0
Fork 0
mirror of https://github.com/pstrueb/piper.git synced 2026-06-02 09:57:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add note for high quality training

Browse Source
This commit is contained in:
Michael Hansen
2023-06-22 11:17:49 -05:00
parent 00c161ab8b
commit fe3652147a
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
TRAINING.md
+2
View File
@@ -170,6 +170,8 @@ python3 -m piper_train \
--precision 32
```
Use `--quality high` to train a [larger voice model](https://github.com/rhasspy/piper/blob/master/src/python/piper_train/vits/config.py#L45) (sounds better, but is much slower).
You can adjust the validation split (5% = 0.05) and number of test examples for your specific dataset. For fine-tuning, they are often set to 0 because the target dataset is very small.
Batch size can be tricky to get right. It depends on the size of your GPU's vRAM, the model's quality/size, and the length of the longest sentence in your dataset. The `--max-phoneme-ids <N>` argument to `piper_train` will drop sentences that have more than `N` phoneme ids. In practice, using `--batch-size 32` and `--max-phoneme-ids 400` will work for 24 GB of vRAM (RTX 3090/4090).