Downloads · 30 days
0
dsa-tokenizer
DiscreteSpeech/dsa-tokenizer
dsa-tokenizer is a text-to-speech model from DiscreteSpeech. Use it when you need text read aloud. The card lists the license as mit.
This repository contains the pretrained model assets for DSA-Tokenizer inference. The code and runnable examples live in the GitHub repository.
Access
Public
Updated Jul 21, 2026
Repo size
13.7 GB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.pt13.7 GB · 100%
At a glance
- Task
- Text-to-Speech
- License
- mit
- Access
- Public
- Created
- Jul 21, 2026
- Updated
- Jul 21, 2026
- SHA
- 78335dab
Model card
From the Hugging Face model README
DSA-Tokenizer
This repository contains the pretrained model assets for DSA-Tokenizer inference. The code and runnable examples live in the GitHub repository.
Files
DiscreteSpeech/dsa-tokenizer
|-- README.md
|-- model_index.json
|-- dsa_tokenizer/
| |-- model_10000.pt
| |-- vocab.txt
| `-- config.yaml
`-- vocoder/
`-- flow2gan_mel_24k_base_50hz/
`-- epoch-20.pt
Usage
from f5_tts.dsa_api import DSATokenizer
dsa = DSATokenizer.from_pretrained("DiscreteSpeech/dsa-tokenizer", device="cuda:0")
- Task
- Text-to-Speech
- License
- mit
- Created
- Jul 21, 2026
- Updated
- Jul 21, 2026