Materials

Links to slides, video, and referenced papers for each individual tutorial section.

Introduction

Task definition

Challenges of speech

Cascaded models

End-to-End

State-of-the-art

Input representations

Architecture & modifications

Output representations

Leveraging Data Sources

Available data

Techniques: Multi-task

Techniques: Transfer learning & pretraining

Techniques: Knowledge distillation

Alternate data representations

Evaluation

Automatic metrics

Utterance segmentation: Evaluation

Mitigating errors: Gender bias

Advanced Topics

Utterance segmentation

Multilingual ST

Under-resourced languages

Real World Applications

Automatic generation of subtitles

Simultaneous translation

Conclusion

Conclusion

References

• , , (editors), , , , , (), pages , , . .
  [bibtex]