Upload 7 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+model filter=lfs diff=lfs merge=lfs -text

D_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:273a1da965da0f3b51c7f630c3aa1bf0ef4739da4ab367a9f063a6e12058e8ce
+size 187027770

G_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da86273856084312fcae6c6adc50f7149baab67693ea9f896117ad20c076dd2e
+size 209268661

README.md

@@ -1,5 +1,43 @@
----
-license: apache-2.0
-license_name: apache2.0
-license_link: LICENSE
----
+---
+license: apache-2.0
+tags:
+- Hibernates
+- HVC-Audio-Convert
+
+pipeline_tag: audio-to-audio
+---
+
+# HVC-Audio-Convert Base Models
+
+## Overview
+These models serve as the foundational components for HVC-Audio-Convert (Soft-VC Voice Conversion), an advanced voice conversion framework that combines SoftVC feature extraction with the VITS (Conditional Variational Autoencoder with Adversarial Learning) architecture.
+
+## Key Features
+- High-quality voice conversion capabilities
+- Pre-trained on diverse vocal datasets
+- Supports cross-lingual voice conversion
+- Compatible with HVC-Audio-Convert v4.0 and newer
+
+## Technical Details
+- **Architecture**: Based on VITS (Conditional Variational Autoencoder)
+- **Feature Extraction**: Hibernates content encoder
+- **Training Data**: Curated multi-speaker datasets
+- **Model Format**: PyTorch checkpoints
+
+## Usage
+1. Download the desired base model
+2. Use with HVC-Audio-Convert framework
+3. Fine-tune on target voice data
+4. Perform voice conversion
+
+## Requirements
+- HVC-Audio-Convert framework
+- Python 3.8+
+- PyTorch 1.13.0+
+- CUDA compatible GPU (recommended)
+
+## License
+This project is licensed under the Apache License 2.0 - see the LICENSE file for details.
+
+## Citation
+If you use these models in your research, please cite:

config.json

@@ -0,0 +1,48 @@
+{
+    "resblock": "2",
+    "num_gpus": 1,
+    "batch_size": 24,
+    "learning_rate": 0.00015,
+    "adam_b1": 0.85,
+    "adam_b2": 0.995,
+    "adam_eps": 1e-8,
+    "lr_decay": 0.9995,
+    "seed": 42,
+
+    "upsample_rates": [6, 6, 4, 2, 2],
+    "upsample_kernel_sizes": [12, 12, 8, 4, 4],
+    "upsample_initial_channel": 384,
+    "resblock_kernel_sizes": [3, 5, 7, 11],
+    "resblock_dilation_sizes": [[1,2,4], [1,2,4], [1,2,4], [1,2,4]],
+    "discriminator_periods": [2, 3, 5, 7, 11, 17, 23, 31],
+
+    "segment_size": 8192,
+    "num_mels": 100,
+    "num_freq": 2049,
+    "n_fft": 4096,
+    "hop_size": 256,
+    "win_size": 4096,
+
+    "sampling_rate": 48000,
+
+    "fmin": 10,
+    "fmax": 24000,
+    "fmax_for_loss": 18000,
+
+    "num_workers": 6,
+
+    "training": {
+        "epochs": 2000,
+        "save_interval": 10,
+        "validation_interval": 2,
+        "grad_clip": 4.0,
+        "warmup_steps": 1000,
+        "decay_steps": 50000
+    },
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54322",
+        "world_size": 1
+    }
+}

model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c576b63b7ed952161b70fad34e0562ace502ce689195520d8a2a6c051de29d6
+size 56825430

model_0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:409452a27ab310f7a5897844d003d372a7357cc91c4a43562584a1714518cdf9
+size 220895384

rmvpe.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19dc1809cf4cdb0a18db93441816bc327e14e5644b72eeaae5220560c6736fe2
+size 368492925