Text-to-Speech
AudioSeal
Safetensors
English
zero-shot
voice-cloning
english
flow-matching
diffusion-transformer
dacvae
Instructions to use VoiceHub/DACFlow-EN-10k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- AudioSeal
How to use VoiceHub/DACFlow-EN-10k with AudioSeal:
# Watermark Generator from audioseal import AudioSeal model = AudioSeal.load_generator("VoiceHub/DACFlow-EN-10k") # pass a tensor (tensor_wav) of shape (batch, channels, samples) and a sample rate wav, sr = tensor_wav, 16000 watermark = model.get_watermark(wav, sr) watermarked_audio = wav + watermark# Watermark Detector from audioseal import AudioSeal detector = AudioSeal.load_detector("VoiceHub/DACFlow-EN-10k") result, message = detector.detect_watermark(watermarked_audio, sr) - Notebooks
- Google Colab
- Kaggle
showcase: refresh the card
Browse files- README.md +1 -1
- showcase.json +15 -2
README.md
CHANGED
|
@@ -176,7 +176,7 @@ UTMOS** are better):
|
|
| 176 |
|
| 177 |
| Checkpoint | echo-dev WER β | echo-dev SIM-o β | echo-dev UTMOS β (real speech: 4.20) | seed-dev WER β | seed-dev SIM-o β | seed-dev UTMOS β (real speech: 3.52) | Samples | Weights |
|
| 178 |
|---|---:|---:|---:|---:|---:|---:|---|---|
|
| 179 |
-
| **step 100k** (newest) |
|
| 180 |
| step 90k | 0.74 % | 0.782 | 3.78 | - | - | - | [listen](#listen) | [files](https://huggingface.co/VoiceHub/DACFlow-EN-10k/tree/main/checkpoints/step_0090000) |
|
| 181 |
| step 80k | 0.78 % | 0.776 | 3.73 | 1.58 % | 0.507 | 3.68 | [listen](#listen) | [files](https://huggingface.co/VoiceHub/DACFlow-EN-10k/tree/main/checkpoints/step_0080000) |
|
| 182 |
| step 70k | 0.67 % | 0.766 | 3.69 | - | - | - | [listen](#listen) | [files](https://huggingface.co/VoiceHub/DACFlow-EN-10k/tree/main/checkpoints/step_0070000) |
|
|
|
|
| 176 |
|
| 177 |
| Checkpoint | echo-dev WER β | echo-dev SIM-o β | echo-dev UTMOS β (real speech: 4.20) | seed-dev WER β | seed-dev SIM-o β | seed-dev UTMOS β (real speech: 3.52) | Samples | Weights |
|
| 178 |
|---|---:|---:|---:|---:|---:|---:|---|---|
|
| 179 |
+
| **step 100k** (newest) | 0.69 % | 0.787 | 3.85 | 1.54 % | 0.499 | 3.78 | [listen](#listen) | [files](https://huggingface.co/VoiceHub/DACFlow-EN-10k/tree/main/checkpoints/step_0100000) |
|
| 180 |
| step 90k | 0.74 % | 0.782 | 3.78 | - | - | - | [listen](#listen) | [files](https://huggingface.co/VoiceHub/DACFlow-EN-10k/tree/main/checkpoints/step_0090000) |
|
| 181 |
| step 80k | 0.78 % | 0.776 | 3.73 | 1.58 % | 0.507 | 3.68 | [listen](#listen) | [files](https://huggingface.co/VoiceHub/DACFlow-EN-10k/tree/main/checkpoints/step_0080000) |
|
| 182 |
| step 70k | 0.67 % | 0.766 | 3.69 | - | - | - | [listen](#listen) | [files](https://huggingface.co/VoiceHub/DACFlow-EN-10k/tree/main/checkpoints/step_0070000) |
|
showcase.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"card_sha": "
|
| 3 |
"checkpoints": {
|
| 4 |
"en_base/step_0020000": {
|
| 5 |
"added": "2026-10-03T02:01:41Z",
|
|
@@ -1038,7 +1038,20 @@
|
|
| 1038 |
},
|
| 1039 |
"folder": "checkpoints/step_0100000",
|
| 1040 |
"lineage": [],
|
| 1041 |
-
"metrics": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1042 |
"note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data",
|
| 1043 |
"params_m": 206.41,
|
| 1044 |
"preset": "base",
|
|
|
|
| 1 |
{
|
| 2 |
+
"card_sha": "94cf5f574da6be42",
|
| 3 |
"checkpoints": {
|
| 4 |
"en_base/step_0020000": {
|
| 5 |
"added": "2026-10-03T02:01:41Z",
|
|
|
|
| 1038 |
},
|
| 1039 |
"folder": "checkpoints/step_0100000",
|
| 1040 |
"lineage": [],
|
| 1041 |
+
"metrics": {
|
| 1042 |
+
"echo_dev": {
|
| 1043 |
+
"n": 260,
|
| 1044 |
+
"sim_o": 0.7872023932062663,
|
| 1045 |
+
"utmos": 3.851715524380024,
|
| 1046 |
+
"wer": 0.006906077348066298
|
| 1047 |
+
},
|
| 1048 |
+
"seed_dev": {
|
| 1049 |
+
"n": 1090,
|
| 1050 |
+
"sim_o": 0.4993639859926263,
|
| 1051 |
+
"utmos": 3.7841195987998892,
|
| 1052 |
+
"wer": 0.01537433155080214
|
| 1053 |
+
}
|
| 1054 |
+
},
|
| 1055 |
"note": "Main model (M4 base pretraining): the 206 M base preset on the full provided data",
|
| 1056 |
"params_m": 206.41,
|
| 1057 |
"preset": "base",
|