Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,31 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: cc-by-4.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-4.0
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
- it
|
| 6 |
+
- pt
|
| 7 |
+
- de
|
| 8 |
+
- fr
|
| 9 |
+
- es
|
| 10 |
+
- ja
|
| 11 |
+
- zh
|
| 12 |
+
tags:
|
| 13 |
+
- automatic-speech-recognition
|
| 14 |
+
- speech
|
| 15 |
+
- audio
|
| 16 |
+
- Transformer
|
| 17 |
+
- flow-matching
|
| 18 |
+
- discrete-flow-matching
|
| 19 |
+
- pytorch
|
| 20 |
+
- hf-asr-leaderboard
|
| 21 |
+
---
|
| 22 |
+
|
| 23 |
+
# Drax: Speech Recognition with Discrete Flow Matching
|
| 24 |
+
|
| 25 |
+
## Model Overview
|
| 26 |
+
|
| 27 |
+
The Drax model family provides speech recognition models based on discrete flow matching.
|
| 28 |
+
The `drax-v1` model supports eight languages: English, Spanish, French, Portuguese, German, Italian, Japanese and Chinese.
|
| 29 |
+
It is an encoder-decoder model consists of a Whisper-large-v3 encoder, and a DiT based decoder, with a total of ~1.2B parameters.
|
| 30 |
+
|
| 31 |
+
More details on usage in our GitHub repo: [https://github.com/aiola-lab/drax]()
|