Add pipeline tag, license, and paper link
#1
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -1,12 +1,16 @@
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
-
|
| 4 |
-
|
|
|
|
|
|
|
| 5 |
---
|
| 6 |
|
| 7 |
# Talk Like You checkpoints
|
| 8 |
|
| 9 |
-
Inference weights for [Talk Like You: Imitating How You Speak in Real-Time Talking Head Generation](https://
|
|
|
|
|
|
|
| 10 |
|
| 11 |
The [source code and usage instructions](https://github.com/BQ-Wang0511/TalkLikeYou) are on GitHub. From the code repository root, download these weights with:
|
| 12 |
|
|
@@ -14,4 +18,4 @@ The [source code and usage instructions](https://github.com/BQ-Wang0511/TalkLike
|
|
| 14 |
hf download doubi-killer/TalkLikeYou --include "checkpoints/**" --local-dir .
|
| 15 |
```
|
| 16 |
|
| 17 |
-
Keep the downloaded directory structure unchanged. Third-party weights retain their original terms. In particular, the bundled InsightFace models are limited to non-commercial research use, and the Wav2Lip-related model has upstream research-use restrictions. See the [third-party notices](https://github.com/BQ-Wang0511/TalkLikeYou/blob/main/THIRD_PARTY_NOTICES.md) before use.
|
|
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
+
- talking-head-generation
|
| 4 |
+
- audio-driven-animation
|
| 5 |
+
license: other
|
| 6 |
+
pipeline_tag: image-to-video
|
| 7 |
---
|
| 8 |
|
| 9 |
# Talk Like You checkpoints
|
| 10 |
|
| 11 |
+
Inference weights for [Talk Like You: Imitating How You Speak in Real-Time Talking Head Generation](https://huggingface.co/papers/2610.06658).
|
| 12 |
+
|
| 13 |
+
Project page: [https://bq-wang0511.github.io/TalkLikeYou/](https://bq-wang0511.github.io/TalkLikeYou/)
|
| 14 |
|
| 15 |
The [source code and usage instructions](https://github.com/BQ-Wang0511/TalkLikeYou) are on GitHub. From the code repository root, download these weights with:
|
| 16 |
|
|
|
|
| 18 |
hf download doubi-killer/TalkLikeYou --include "checkpoints/**" --local-dir .
|
| 19 |
```
|
| 20 |
|
| 21 |
+
Keep the downloaded directory structure unchanged. Third-party weights retain their original terms. In particular, the bundled InsightFace models are limited to non-commercial research use, and the Wav2Lip-related model has upstream research-use restrictions. See the [third-party notices](https://github.com/BQ-Wang0511/TalkLikeYou/blob/main/THIRD_PARTY_NOTICES.md) before use.
|