|
|
| 第70行： |
第70行： |
| | |Zixi Yan | | |Zixi Yan |
| | || | | || |
| − | * | + | * Read SSAST: Self-Supervised Audio Spectrogram Transformer |
| | + | * Wav2vec multi-layer feature combination research and system design |
| | || | | || |
| | * | | * |
| People |
This Week |
Next Week |
Task Tracking (DeadLine)
|
| Dong Wang
|
- AI-graph online resource
- Submit PAMI paper
- Reformate hard-trials paper
- Formate CNSRC paper
|
|
|
| Lantian Li
|
|
|
|
| Ying Shi
|
- coding
- aishell course video
|
- finish coding
- transformer keyword spotting baseline
- instance keyword baseline
|
|
| Chen Chen
|
- doc
- Finish data_collector version 0.1
- Processed two mini batch
|
- Manually evaluate generated videos and finetune data_collector
|
|
| Pengqi Li
|
- study some post-hoc method
|
|
|
| Zixi Yan
|
- Read SSAST: Self-Supervised Audio Spectrogram Transformer
- Wav2vec multi-layer feature combination research and system design
|
|
|
| Sirui Li
|
- sv task: use wav2vec hubert mfcc feature
|
|
|
| Haoyu Jiang
|
|
|
|
| Ruihai Hou
|
|
|
|
| Renmiao Chen
|
- Determine our plan
- Finish audio_drived_data_collector version 0.1
- Anaylze some example
- Design strategies for collection
- conclusion & preliminary ideas
|
- Finish video-drived collection data collector
|
|
| Wenqiang Du
|
|
- Some project cooperation
- English acoustic model upgrade
|
|
| Wan Lin
|
- Test the method mentioned in the last report
- Papers on noise filtering
|
- Find ways to improve model adaptability
|
|
