Updated fine-tuned 44k singing voice conversion model with better audio quality

Author: Plachtaa

README.md

@@ -205,10 +205,12 @@ Generally, latency is around 1~2s to prevent quality drop (the sad nature of dif
 - [x] Changed to BigVGAN from NVIDIA for singing voice decoding
 - [x] Whisper version model for singing voice conversion
 - [x] Objective evaluation and comparison with RVC/SoVITS for singing voice conversion
-- [ ] Improve audio quality
+- [x] Improve audio quality
 - [ ] More to be added
 
 ## CHANGELOGS🗒️
+- 2024-10-28:
+    - Updated fine-tuned 44k singing voice conversion model with better audio quality
 - 2024-10-27:
     - Added real-time voice conversion GUI
 - 2024-10-25:

app.py

@@ -84,7 +84,7 @@
 
 # f0 conditioned model
 dit_checkpoint_path, dit_config_path = load_custom_model_from_hf("Plachta/Seed-VC",
-                                                "DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ema.pth",
+                                                "DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ft_ema.pth",
                                                 "config_dit_mel_seed_uvit_whisper_base_f0_44k.yml")
 
 config = yaml.safe_load(open(dit_config_path, 'r'))

inference.py

@@ -31,7 +31,7 @@ def load_models(args):
                                                                          "config_dit_mel_seed_uvit_whisper_small_wavenet.yml")
     else:
         dit_checkpoint_path, dit_config_path = load_custom_model_from_hf("Plachta/Seed-VC",
-                                                                        "DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ema.pth",
+                                                                        "DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ema_ft.pth",
                                                                         "config_dit_mel_seed_uvit_whisper_base_f0_44k.yml")
         # f0 extractor
         from modules.rmvpe import RMVPE