AutoMode-ASR: Learning to Select ASR Systems for Better Quality and Cost
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909319702773760 |
|---|---|
| author | Gündüz, Ahmet Kim, Yunsu Yuksel, Kamer Ali Al-Badrashiny, Mohamed Ferreira, Thiago Castro Sawaf, Hassan |
| author_facet | Gündüz, Ahmet Kim, Yunsu Yuksel, Kamer Ali Al-Badrashiny, Mohamed Ferreira, Thiago Castro Sawaf, Hassan |
| contents | We present AutoMode-ASR, a novel framework that effectively integrates multiple ASR systems to enhance the overall transcription quality while optimizing cost. The idea is to train a decision model to select the optimal ASR system for each segment based solely on the audio input before running the systems. We achieve this by ensembling binary classifiers determining the preference between two systems. These classifiers are equipped with various features, such as audio embeddings, quality estimation, and signal properties. Additionally, we demonstrate how using a quality estimator can further improve performance with minimal cost increase. Experimental results show a relative reduction in WER of 16.2%, a cost saving of 65%, and a speed improvement of 75%, compared to using a single-best model for all segments. Our framework is compatible with commercial and open-source black-box ASR systems as it does not require changes in model codes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_12476 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | AutoMode-ASR: Learning to Select ASR Systems for Better Quality and Cost Gündüz, Ahmet Kim, Yunsu Yuksel, Kamer Ali Al-Badrashiny, Mohamed Ferreira, Thiago Castro Sawaf, Hassan Computation and Language Sound Audio and Speech Processing We present AutoMode-ASR, a novel framework that effectively integrates multiple ASR systems to enhance the overall transcription quality while optimizing cost. The idea is to train a decision model to select the optimal ASR system for each segment based solely on the audio input before running the systems. We achieve this by ensembling binary classifiers determining the preference between two systems. These classifiers are equipped with various features, such as audio embeddings, quality estimation, and signal properties. Additionally, we demonstrate how using a quality estimator can further improve performance with minimal cost increase. Experimental results show a relative reduction in WER of 16.2%, a cost saving of 65%, and a speed improvement of 75%, compared to using a single-best model for all segments. Our framework is compatible with commercial and open-source black-box ASR systems as it does not require changes in model codes. |
| title | AutoMode-ASR: Learning to Select ASR Systems for Better Quality and Cost |
| topic | Computation and Language Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2409.12476 |