Add paper, GitHub and Unified-TS-ASR model links
Browse files
README.md
CHANGED
|
@@ -22,6 +22,9 @@ cue and ignores everyone else. If nobody says the cue, it rejects the request.
|
|
| 22 |
- Demo: [Unified-TS-ASR Space](https://huggingface.co/spaces/YuCeong-May/Unified-TS-ASR)
|
| 23 |
(select "Text-Cue-Guided TS-ASR").
|
| 24 |
- Inference code: the `tcg_tsasr` package in the Space repository.
|
|
|
|
|
|
|
|
|
|
| 25 |
- Language: Mandarin Chinese, 16 kHz audio. Audio within 20 seconds is recommended.
|
| 26 |
|
| 27 |
## Model
|
|
|
|
| 22 |
- Demo: [Unified-TS-ASR Space](https://huggingface.co/spaces/YuCeong-May/Unified-TS-ASR)
|
| 23 |
(select "Text-Cue-Guided TS-ASR").
|
| 24 |
- Inference code: the `tcg_tsasr` package in the Space repository.
|
| 25 |
+
- Paper: [Unified Target-Speaker ASR with Text and Enrollment Speech Cues](https://arxiv.org/abs/2609.33853) (arXiv:2609.33853)
|
| 26 |
+
- Code: [github.com/YuCeong-May/Unified-TS-ASR](https://github.com/YuCeong-May/Unified-TS-ASR)
|
| 27 |
+
- Unified-TS-ASR model: [`YuCeong-May/Unified-TS-ASR`](https://huggingface.co/YuCeong-May/Unified-TS-ASR)
|
| 28 |
- Language: Mandarin Chinese, 16 kHz audio. Audio within 20 seconds is recommended.
|
| 29 |
|
| 30 |
## Model
|