update structure

README.md

@@ -135,16 +135,11 @@ pip install -r requirements.txt
 ```
 minimax-speech/
 ├── assets/
-│   └── image.png
-├── configs/
-│   └── dac_vae.yaml
-├── models/
-│   ├── fsq/
-│   └── dac_vae/
-├── cosyvoice/          # Components from CosyVoice2
+├── dac-vae/
+├── flowae/
+├── speech/    
+│   ├── llm/
 │   ├── flow/
-│   ├── transformer/
-│   └── utils/
 └── README.md
 ```
 

speech/config.yaml

@@ -198,7 +198,7 @@ sort: !name:cosyvoice.dataset.processor.sort
     sort_size: 500  # sort_size should be less than shuffle_size
 batch: !name:cosyvoice.dataset.processor.batch
     batch_type: 'dynamic'
-    max_frames_in_batch: 50000
+    max_frames_in_batch: 40000
 padding: !name:cosyvoice.dataset.processor.padding
     use_spk_embedding: False # change to True during sft
     use_speaker_encoder: !ref <use_speaker_encoder>