Update README.md
Browse files
README.md
CHANGED
|
@@ -5,7 +5,7 @@ license: cc-by-nc-nd-4.0
|
|
| 5 |
# ConsistencyTTA: Accelerating Diffusion-Based Text-to-Audio Generation with Consistency Distillation
|
| 6 |
|
| 7 |
This page shares the official model checkpoints of the paper \
|
| 8 |
-
|
| 9 |
from Microsoft Applied Science Group and UC Berkeley \
|
| 10 |
by [Yatong Bai](https://bai-yt.github.io),
|
| 11 |
[Trung Dang](https://www.microsoft.com/applied-sciences/people/trung-dang),
|
|
@@ -13,6 +13,7 @@ by [Yatong Bai](https://bai-yt.github.io),
|
|
| 13 |
[Kazuhito Koishida](https://www.microsoft.com/applied-sciences/people/kazuhito-koishida),
|
| 14 |
and [Somayeh Sojoudi](https://people.eecs.berkeley.edu/~sojoudi/).
|
| 15 |
|
|
|
|
| 16 |
**[[Preprint Paper](https://arxiv.org/abs/2309.10740)]**
|
| 17 |
**[[Project Homepage](https://consistency-tta.github.io)]**
|
| 18 |
**[[Code](https://github.com/Bai-YT/ConsistencyTTA)]**
|
|
@@ -22,6 +23,11 @@ and [Somayeh Sojoudi](https://people.eecs.berkeley.edu/~sojoudi/).
|
|
| 22 |
|
| 23 |
## Description
|
| 24 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
This work proposes a *consistency distillation* framework to train
|
| 26 |
text-to-audio (TTA) generation models that only require a single neural network query,
|
| 27 |
reducing the computation of the core step of diffusion-based TTA models by a factor of 400.
|
|
|
|
| 5 |
# ConsistencyTTA: Accelerating Diffusion-Based Text-to-Audio Generation with Consistency Distillation
|
| 6 |
|
| 7 |
This page shares the official model checkpoints of the paper \
|
| 8 |
+
*ConsistencyTTA: Accelerating Diffusion-Based Text-to-Audio Generation with Consistency Distillation* \
|
| 9 |
from Microsoft Applied Science Group and UC Berkeley \
|
| 10 |
by [Yatong Bai](https://bai-yt.github.io),
|
| 11 |
[Trung Dang](https://www.microsoft.com/applied-sciences/people/trung-dang),
|
|
|
|
| 13 |
[Kazuhito Koishida](https://www.microsoft.com/applied-sciences/people/kazuhito-koishida),
|
| 14 |
and [Somayeh Sojoudi](https://people.eecs.berkeley.edu/~sojoudi/).
|
| 15 |
|
| 16 |
+
**[[🤗 Live Demo](https://huggingface.co/spaces/Bai-YT/ConsistencyTTA)]**
|
| 17 |
**[[Preprint Paper](https://arxiv.org/abs/2309.10740)]**
|
| 18 |
**[[Project Homepage](https://consistency-tta.github.io)]**
|
| 19 |
**[[Code](https://github.com/Bai-YT/ConsistencyTTA)]**
|
|
|
|
| 23 |
|
| 24 |
## Description
|
| 25 |
|
| 26 |
+
**2024/06 Updates:**
|
| 27 |
+
|
| 28 |
+
- We have hosted an interactive live demo of ConsistencyTTA at [🤗 Huggingface](https://huggingface.co/spaces/Bai-YT/ConsistencyTTA).
|
| 29 |
+
- ConsistencyTTA has been accepted to ***INTERSPEECH 2024***! We look forward to meeting you in Kos Island.
|
| 30 |
+
|
| 31 |
This work proposes a *consistency distillation* framework to train
|
| 32 |
text-to-audio (TTA) generation models that only require a single neural network query,
|
| 33 |
reducing the computation of the core step of diffusion-based TTA models by a factor of 400.
|