Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.20860 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914111101599744 |
|---|---|
| author | Udandarao, Vishaal Lu, Zhiyun Chang, Xuankai Wang, Yongqiang Yao, Violet Z. Jose, Albin Madapally Faghri, Fartash Gardner, Josh Chiu, Chung-Cheng |
| author_facet | Udandarao, Vishaal Lu, Zhiyun Chang, Xuankai Wang, Yongqiang Yao, Violet Z. Jose, Albin Madapally Faghri, Fartash Gardner, Josh Chiu, Chung-Cheng |
| contents | Spoken Question-Answering (SQA) is a core capability for useful and interactive artificial intelligence systems. Recently, several speech-language models (SpeechLMs) have been released with a specific focus on improving their SQA performance. However, a lack of controlled ablations of pretraining data processing and curation makes it challenging to understand what factors account for performance, despite substantial gains from similar studies in other data modalities. In this work, we address this gap by conducting a data-centric exploration for pretraining SpeechLMs. We focus on three research questions fundamental to speech-language pretraining data: (1) how to process raw web-crawled audio content for speech-text pretraining, (2) how to construct synthetic pretraining datasets to augment web-crawled data and (3) how to interleave (text, audio) segments into training sequences. We apply the insights from our controlled data-centric ablations to pretrain a 3.8B-parameter SpeechLM, called SpeLangy, that outperforms models that are up to 3x larger by 10.2% absolute performance. We hope our findings highlight the impact of effective data curation for speech-language pretraining and guide future data-centric exploration in SpeechLMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_20860 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Data-Centric Lessons To Improve Speech-Language Pretraining Udandarao, Vishaal Lu, Zhiyun Chang, Xuankai Wang, Yongqiang Yao, Violet Z. Jose, Albin Madapally Faghri, Fartash Gardner, Josh Chiu, Chung-Cheng Audio and Speech Processing Computation and Language Machine Learning Spoken Question-Answering (SQA) is a core capability for useful and interactive artificial intelligence systems. Recently, several speech-language models (SpeechLMs) have been released with a specific focus on improving their SQA performance. However, a lack of controlled ablations of pretraining data processing and curation makes it challenging to understand what factors account for performance, despite substantial gains from similar studies in other data modalities. In this work, we address this gap by conducting a data-centric exploration for pretraining SpeechLMs. We focus on three research questions fundamental to speech-language pretraining data: (1) how to process raw web-crawled audio content for speech-text pretraining, (2) how to construct synthetic pretraining datasets to augment web-crawled data and (3) how to interleave (text, audio) segments into training sequences. We apply the insights from our controlled data-centric ablations to pretrain a 3.8B-parameter SpeechLM, called SpeLangy, that outperforms models that are up to 3x larger by 10.2% absolute performance. We hope our findings highlight the impact of effective data curation for speech-language pretraining and guide future data-centric exploration in SpeechLMs. |
| title | Data-Centric Lessons To Improve Speech-Language Pretraining |
| topic | Audio and Speech Processing Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2510.20860 |