123

README.md

@@ -3,14 +3,13 @@ language:
 - en
 - zh
 pipeline_tag: text-to-audio
-library_name: tencent-song-generation
+library_name: transformers
 ---
 
 # SongGeneration
 
-<p align="center"><img src="img/logo.jpg" width="40%"></p>
 <p align="center">
-    <a href="https://levo-demo.github.io/">Demo</a> &nbsp;|&nbsp; <a href="https://arxiv.org/abs/2506.07520">Paper</a>  &nbsp;|&nbsp; <a href="https://github.com/tencent-ailab/songgeneration">Code</a>  &nbsp;|&nbsp; <a href="https://huggingface.co/spaces/tencent/SongGeneration">Space Demo</a>
+    <a href="https://levo-demo.github.io/">Demo</a> &nbsp;|&nbsp; <a href="https://arxiv.org/abs/2506.07520">Paper</a>  &nbsp;|&nbsp; <a href="https://github.com/tencent-ailab/songgeneration">Code</a>  &nbsp;|&nbsp; <a href="https://huggingface.co/spaces/waytan22/SongGeneration-LeVo">Space Demo</a>
 </p>
 
 
@@ -20,7 +19,7 @@ This repository is the official weight repository for LeVo: High-Quality Song Ge
 
 |          Model           |                         HuggingFace                          |
 | :----------------------: | :----------------------------------------------------------: |
-|  SongGeneration-base   | <a href="https://huggingface.co/tencent/SongGeneration/tree/main/ckpt/songgeneration_base">v20250520</a> |
+|  SongGeneration-base(zh)   | <a href="https://huggingface.co/tencent/SongGeneration/tree/main/ckpt/songgeneration_base_zh">v20250520</a> |
 | SongGeneration-base(zh&en) |                         Coming soon                          |
 | SongGeneration-full(zh&en) |                         Coming soon                          |
 
@@ -32,4 +31,4 @@ We develop the SongGeneration model. It is an LM-based framework consisting of *
 
 ## License
 
-The code and weights in this repository is released in the [LICENSE](LICENSE)  file.
+The code and weights in this repository is released in the [LICENSE](LICENSE)  file.

ckpt/model_septoken/model_2.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:430b7c1c245722fbe3893cd621b3d4a90076404596e9fb1ce987a4a0f2a4fc6f
+oid sha256:758aa342942a7b7c0ae179af1a952e0b944e39128ea816741499b3031113aaee
 size 4808167708

ckpt/{songgeneration_base → songgeneration_base_zh}/config.yaml

@@ -106,36 +106,3 @@ conditioners:
     QwTextTokenizer:
       token_path: third_party/Qwen2-7B
       max_len: 50
-
-offload:
-  audiolm:
-    offload_module: self
-    cpu_mem_gb: 0
-    pre_copy_step: 1
-    clean_cache_after_forward: false
-    dtype: torch.float16
-    offload_layer_dict:
-      transformer: 4
-      transformer2: 4
-    ignore_layer_list: []
-    clean_cache_wrapper:
-      module: self
-      method_name: _sample_next_token
-      diff_mem_gb_thre: 2
-    debug: false
-
-  wav_tokenizer_diffusion:
-    offload_module: self.model.model
-    pre_copy_step: 1
-    clean_cache_after_forward: false
-    cpu_mem_gb: -1
-    dtype: null
-    offload_layer_dict:
-      cfm_wrapper: 5
-      hubert: 4
-    ignore_layer_list: []
-    clean_cache_wrapper:
-      module: self.model.model.cfm_wrapper.estimator
-      method_name: forward
-      diff_mem_gb_thre: 1
-    debug: false