Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags: [resonance, anima, tts, speech, from-scratch]
|
| 4 |
+
library_name: fsp-ttv
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
# FatStinkyPanda · Resonance — step 5,000
|
| 8 |
+
|
| 9 |
+
A from-scratch **neural text-to-speech** engine (~309M params, SNAC 24 kHz codec), trained in the **Anima** project (https://github.com/FatStinkyPanda/Anima). Created & released by **FatStinkyPanda** (Daniel A. Bissey).
|
| 10 |
+
|
| 11 |
+
> Honest scope: a custom architecture trained on a tiny budget. This repo holds the raw checkpoint — it loads with the project's `fsp-ttv` engine, NOT `transformers`.
|
| 12 |
+
|
| 13 |
+
## Synthesize (fsp-ttv engine)
|
| 14 |
+
```bash
|
| 15 |
+
pip install 'git+https://github.com/FatStinkyPanda/Anima#subdirectory=engine'
|
| 16 |
+
fsp-ttv synthesize "Hello from Resonance." --ckpt final.pt -o out.wav
|
| 17 |
+
```
|
| 18 |
+
|
| 19 |
+
## Latest validation health scores (0..1, from the engine's own instrument)
|
| 20 |
+
- **behavior**: 1.0
|
| 21 |
+
- **layer_health**: 1.0
|
| 22 |
+
- **stability**: 1.0
|
| 23 |
+
- **coarse_accuracy**: 0.0042
|
| 24 |
+
- **overall**: 0.6384
|
| 25 |
+
- **tf_coarse_accuracy**: 0.0042
|