CN-CVS - 版本历史

2022年10月30日 (日) 11:47 Cchen

2022-10-30T11:47:21Z

Cchen：Cchen移动页面MVS至CN-CVS

2022-10-30T11:44:24Z

Cchen移动页面MVS至CN-CVS

Cchen：/* Source Code */

2022-10-25T06:24:21Z

‎Source Code

Cchen：/* Future Plans */

2022-10-25T05:09:15Z

‎Future Plans

Cchen：以“===Introduction=== * Mandarin Visual Speech, a large-scale Chinese Mandarin audio-visual dataset published by Center for Speech and Language Technology (CSLT) at Ts...”为内容创建页面

2022-10-23T09:49:54Z

以“===Introduction=== * Mandarin Visual Speech, a large-scale Chinese Mandarin audio-visual dataset published by Center for Speech and Language Technology (CSLT) at Ts...”为内容创建页面

新页面

===Introduction===

* Mandarin Visual Speech, a large-scale Chinese Mandarin audio-visual dataset published by Center for Speech and Language Technology (CSLT) at Tsinghua University.

===Members===

* Current：Dong Wang, Chen Chen

===Description===

* Collect audio and video data of more than 2500 Mandarin speakers.
* Automatically clip videos through a pipeline including shot detection, VAD, face detection, face tracker, audio-visual synchronization detection.
* Manually annotate speaker identity, human check data quality.
* Create a benchmark database for video to speech synthesis task.

===Basic Methods===

* Environments: PyTorch, OpenCV, FFmpeg
* Shot detection: ffmpeg
* VAD: pydub
* Face detection and tracking: dlib.
* Audio-visual synchronization detection: SyncNet model.
* Input: json files of video information.
* Output: videos clips and wav files, as well as metadata json files.

===Reports===

===Publications===
<pre>

</pre>
===Source Code===

* Collection Pipeline: TODO

===Download===

* Public (recommended)
TODO

* Local (not recommended)
TODO

===Future Plans===

===License===

* All the resources contained in the database are free for research institutes and individuals.
* <b>No commerical usage is permitted</b>.

===References===

@@ 第1行： / 第1行： @@
 ===Introduction===
-* Mandarin Visual Speech, a large-scale Chinese Mandarin audio-visual dataset published by Center for Speech and Language Technology (CSLT) at Tsinghua University.
+* CN-CVS, a large-scale Chinese Mandarin audio-visual dataset published by Center for Speech and Language Technology (CSLT) at Tsinghua University.
 ===Members===
@@ 第32行： / 第32行： @@
 ===Source Code===
-* Collection Pipeline: https://github.com/sectum1919/mvs_data_collector
+* Collection Pipeline: https://github.com/sectum1919/cncvs_data_collector
 * xTS: TODO
 * VCA-GAN: TODO
@@ 第39行： / 第39行： @@
 * Public (recommended)
-TODO
+https://cloud.tsinghua.edu.cn/d/83f13126daec49deb8a3/
 * Local (not recommended)
-TODO
+https://cloud.tsinghua.edu.cn/d/83f13126daec49deb8a3/
 ===Future Plans===
 * Extract text transcription via OCR & ASR & Human check
 ===License===

←上一版本		2022年10月25日 (二) 05:09的版本
第43行：		第43行：

	===Future Plans===		===Future Plans===
		+	* Extract text transcription via OCR & ASR & Human check

	===License===		===License===

←上一版本	2022年10月30日 (日) 11:44的版本
（没有差异）