Restore centroid demo artifacts
Browse filesRestore small checkpoint centroid demo artifacts and document centroid-only demo usage. No training data, optimizer state, scheduler state, logs, generated samples, or local configs are added.
- README.md +28 -0
- USAGE.md +14 -0
- checkpoints/hung_yi_lee_speaker_centroids.pt +0 -0
- checkpoints/speaker_centroids.pt +0 -0
README.md
CHANGED
|
@@ -85,6 +85,33 @@ to synthesize. Reference clips should be at least 3 seconds. The
|
|
| 85 |
speaker-reference-audio-only mode uses the raw reference-audio token path and
|
| 86 |
does not require a reference transcript.
|
| 87 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
## Long Text
|
| 89 |
|
| 90 |
For long-form synthesis, split text into sentence-sized chunks and concatenate
|
|
@@ -127,4 +154,5 @@ Long-form diagnostics on the selected checkpoint:
|
|
| 127 |
- `audiovae.pth`: AudioVAE weights.
|
| 128 |
- `config.json`: BlueMagpie architecture/runtime config.
|
| 129 |
- `tokenizer.json`, `tokenizer_config.json`: tokenizer files.
|
|
|
|
| 130 |
- `USAGE.md`: expanded usage guide.
|
|
|
|
| 85 |
speaker-reference-audio-only mode uses the raw reference-audio token path and
|
| 86 |
does not require a reference transcript.
|
| 87 |
|
| 88 |
+
## Centroid Demo
|
| 89 |
+
|
| 90 |
+
This repository keeps two small centroid demo artifacts under `checkpoints/` so
|
| 91 |
+
users can test the `centroid-only` path without building a centroid table first:
|
| 92 |
+
|
| 93 |
+
- `checkpoints/hung_yi_lee_speaker_centroids.pt`
|
| 94 |
+
- `checkpoints/speaker_centroids.pt`
|
| 95 |
+
|
| 96 |
+
These files are demo speaker-conditioning artifacts only. They are not training
|
| 97 |
+
data, optimizer state, scheduler state, logs, generated samples, or a complete
|
| 98 |
+
speaker registry.
|
| 99 |
+
|
| 100 |
+
```python
|
| 101 |
+
import torch
|
| 102 |
+
|
| 103 |
+
table = torch.load("checkpoints/speaker_centroids.pt", map_location="cpu")
|
| 104 |
+
speaker_ids = table["speaker_ids"]
|
| 105 |
+
centroids = table["centroids"]
|
| 106 |
+
|
| 107 |
+
audio = model.generate(
|
| 108 |
+
target_text="這是 centroid-only 的測試。",
|
| 109 |
+
speaker_centroid=centroids[0],
|
| 110 |
+
cfg_value=2.0,
|
| 111 |
+
inference_timesteps=10,
|
| 112 |
+
)
|
| 113 |
+
```
|
| 114 |
+
|
| 115 |
## Long Text
|
| 116 |
|
| 117 |
For long-form synthesis, split text into sentence-sized chunks and concatenate
|
|
|
|
| 154 |
- `audiovae.pth`: AudioVAE weights.
|
| 155 |
- `config.json`: BlueMagpie architecture/runtime config.
|
| 156 |
- `tokenizer.json`, `tokenizer_config.json`: tokenizer files.
|
| 157 |
+
- `checkpoints/*speaker_centroids.pt`: small centroid demo artifacts.
|
| 158 |
- `USAGE.md`: expanded usage guide.
|
USAGE.md
CHANGED
|
@@ -55,6 +55,20 @@ wav = model.generate(
|
|
| 55 |
|
| 56 |
```
|
| 57 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
Streaming:
|
| 59 |
|
| 60 |
```python
|
|
|
|
| 55 |
|
| 56 |
```
|
| 57 |
|
| 58 |
+
Centroid demo:
|
| 59 |
+
|
| 60 |
+
```python
|
| 61 |
+
import torch
|
| 62 |
+
|
| 63 |
+
table = torch.load(f"{model_dir}/checkpoints/speaker_centroids.pt", map_location="cpu")
|
| 64 |
+
wav = model.generate(
|
| 65 |
+
target_text="這是 centroid demo 的測試。",
|
| 66 |
+
speaker_centroid=table["centroids"][0],
|
| 67 |
+
cfg_value=2.0,
|
| 68 |
+
inference_timesteps=10,
|
| 69 |
+
)
|
| 70 |
+
```
|
| 71 |
+
|
| 72 |
Streaming:
|
| 73 |
|
| 74 |
```python
|
checkpoints/hung_yi_lee_speaker_centroids.pt
ADDED
|
Binary file (2.69 kB). View file
|
|
|
checkpoints/speaker_centroids.pt
ADDED
|
Binary file (3.31 kB). View file
|
|
|