serena checkpoint epoch 94 (medium, base per finetuning italiano)

README.md

@@ -0,0 +1,43 @@
+# it_IT-serena — Piper training checkpoint (Italian, medium)
+
+Finetuning base checkpoint for Piper TTS (VITS), trained **from scratch** on Italian.
+
+- Model: piper (piper_train), quality **medium**, 22.05 kHz
+- Base voice: **serena** — female, neutral Standard Italian (no regional accent), synthetic
+- Training: **94 epochs** from scratch, batch 10, ~29,470 clips (~27 h)
+- Dataset: [committa/serena-synthetic-it-27h](https://huggingface.co/datasets/committa/serena-synthetic-it-27h)
+- File: `epoch=94-step=559930.ckpt` — **keep this filename**: tooling (e.g. the TextyMcSpeechy dojo) parses the `epoch=N-step=M` pattern to pick the highest checkpoint
+
+## Why this checkpoint
+
+Piper publishes **no official pretrained checkpoint for Italian** (the `rhasspy/piper-checkpoints`
+catalog has no `it` entry; the only official Italian voices are `it_IT-paola-medium`, which was
+finetuned from a U.S. English base, and `it_IT-riccardo-x_low`, trained from scratch at x_low
+quality — neither exposes a trainable checkpoint). This is the only public Italian **medium**
+checkpoint that can be used as a finetuning base.
+
+## Use as a base for finetuning other Italian voices
+
+Any checkpoint of the same quality config (medium) can be resumed with piper_train:
+
+```sh
+python -m piper_train \
+    --dataset-dir /path/to/your_voice_training_folder \
+    --resume_from_checkpoint epoch=94-step=559930.ckpt \
+    --quality medium
+```
+
+Because the base already speaks Italian (correct phonemes and prosody), a new Italian voice
+converges in **far fewer epochs** than from scratch — and with less data. This works for
+**both female and male target voices**: the timbre is learned from the new dataset, so with
+enough epochs the new voice fully takes over.
+
+### TextyMcSpeechy dojo
+
+Copy the `.ckpt` into `tts_dojo/PRETRAINED_CHECKPOINTS/default/F_voice/medium/` (or the dojo's
+`pretrained_tts_checkpoint/` folder) and select the "pretrained checkpoint" option in `train.sh`.
+
+## License
+
+CC-BY-4.0 — derived from [committa/serena-synthetic-it-27h](https://huggingface.co/datasets/committa/serena-synthetic-it-27h).
+If you use this checkpoint to build a voice, credit the dataset and this model.

epoch=94-step=559930.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e20efaaf795520b01e4fca6088d58040cd27da28f30e6fab0778d2b9d27d781
+size 845898456