Abstract
The two most common paradigms for end-to-end speech recognition are connectionist temporal classification (CTC) and attention-based encoder-decoder (AED) models. It has been argued that the latter is better suited for learning an implicit language model. We test this hypothesis by measuring temporal context sensitivity and evaluate how the models perform when we constrain the amount of contextual information in the audio input. We find that the AED model is indeed more context sensitive, but that the gap can be closed by adding self-attention to the CTC model. Furthermore, the two models perform similarly when contextual information is constrained. Finally, in contrast to previous research, our results show that the CTC model is highly competitive on WSJ and LibriSpeech without the help of an external language model.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the Annual Conference of the International Speech Communication Association |
| Publication date | 2020 |
| Pages | 4352-4356 |
| DOIs | |
| Publication status | Published - 2020 |
| Event | Interspeech 2020 - Shanghai International Convention Center, Shanghai, China Duration: 25 Oct 2020 → 29 Oct 2020 http://www.interspeech2020.org/ |
Conference
| Conference | Interspeech 2020 |
|---|---|
| Location | Shanghai International Convention Center |
| Country/Territory | China |
| City | Shanghai |
| Period | 25/10/2020 → 29/10/2020 |
| Internet address |
| Series | Proceedings of the Annual Conference of the International Speech Communication Association, Interspeech |
|---|---|
| ISSN | 1990-9772 |
Keywords
- Attention-based encoder-decoder
- Automatic speech recognition
- Connectionist temporal classification
- End-to-end speech recognition