Synth4Kws: Synthesized Speech for User Defined Keyword Spotting in Low Resource Environments
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_ | 1866910011820605440 |
|---|---|
| author | Zhu, Pai Agarwal, Dhruuv Bartel, Jacob W. Partridge, Kurt Park, Hyun Jin Wang, Quan |
| author_facet | Zhu, Pai Agarwal, Dhruuv Bartel, Jacob W. Partridge, Kurt Park, Hyun Jin Wang, Quan |
| contents | One of the challenges in developing a high quality custom keyword spotting (KWS) model is the lengthy and expensive process of collecting training data covering a wide range of languages, phrases and speaking styles. We introduce Synth4Kws - a framework to leverage Text to Speech (TTS) synthesized data for custom KWS in different resource settings. With no real data, we found increasing TTS phrase diversity and utterance sampling monotonically improves model performance, as evaluated by EER and AUC metrics over 11k utterances of the speech command dataset. In low resource settings, with 50k real utterances as a baseline, we found using optimal amounts of TTS data can improve EER by 30.1% and AUC by 46.7%. Furthermore, we mix TTS data with varying amounts of real data and interpolate the real data needed to achieve various quality targets. Our experiments are based on English and single word utterances but the findings generalize to i18n languages and other keyword types. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_16840 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Synth4Kws: Synthesized Speech for User Defined Keyword Spotting in Low Resource Environments Zhu, Pai Agarwal, Dhruuv Bartel, Jacob W. Partridge, Kurt Park, Hyun Jin Wang, Quan Audio and Speech Processing Artificial Intelligence One of the challenges in developing a high quality custom keyword spotting (KWS) model is the lengthy and expensive process of collecting training data covering a wide range of languages, phrases and speaking styles. We introduce Synth4Kws - a framework to leverage Text to Speech (TTS) synthesized data for custom KWS in different resource settings. With no real data, we found increasing TTS phrase diversity and utterance sampling monotonically improves model performance, as evaluated by EER and AUC metrics over 11k utterances of the speech command dataset. In low resource settings, with 50k real utterances as a baseline, we found using optimal amounts of TTS data can improve EER by 30.1% and AUC by 46.7%. Furthermore, we mix TTS data with varying amounts of real data and interpolate the real data needed to achieve various quality targets. Our experiments are based on English and single word utterances but the findings generalize to i18n languages and other keyword types. |
| title | Synth4Kws: Synthesized Speech for User Defined Keyword Spotting in Low Resource Environments |
| topic | Audio and Speech Processing Artificial Intelligence |
| url | https://arxiv.org/abs/2407.16840 |