primepake commited on
Commit
24941fa
·
1 Parent(s): 3595b78

add fsq training

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -11,6 +11,7 @@ This repository provides an implementation of the MiniMax-Speech model, featurin
11
  ## Key Features
12
 
13
  - [ ] **24kHz Audio Support**: High-quality audio generation at 24kHz sampling rate
 
14
  - [ ] **Two-Stage Architecture**: Optimized training pipeline with discrete and continuous representations
15
  - [ ] **Modular Design**: Separate components for audio codec and variational autoencoder
16
  - [ ] **CosyVoice2 Decoder**: Leverages proven components from the CosyVoice2's Decoder framework
 
11
  ## Key Features
12
 
13
  - [ ] **24kHz Audio Support**: High-quality audio generation at 24kHz sampling rate
14
+ - [ ] **FSQ tokenizer training**: Training FSQ from scratch
15
  - [ ] **Two-Stage Architecture**: Optimized training pipeline with discrete and continuous representations
16
  - [ ] **Modular Design**: Separate components for audio codec and variational autoencoder
17
  - [ ] **CosyVoice2 Decoder**: Leverages proven components from the CosyVoice2's Decoder framework