Evaluation of Speech Foundation Models for ASR on Child-Adult Conversations in Autism Diagnostic Sessions
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909736148926464 |
|---|---|
| author | Ashvin, Aditya Lahiri, Rimita Kommineni, Aditya Bishop, Somer Lord, Catherine Kadiri, Sudarsana Reddy Narayanan, Shrikanth |
| author_facet | Ashvin, Aditya Lahiri, Rimita Kommineni, Aditya Bishop, Somer Lord, Catherine Kadiri, Sudarsana Reddy Narayanan, Shrikanth |
| contents | Reliable transcription of child-adult conversations in clinical settings is crucial for diagnosing developmental disorders like Autism. Recent advances in deep learning and availability of large scale transcribed data has led to development of speech foundation models that have shown dramatic improvements in ASR performance. However, their performance on conversational child-adult interactions remains underexplored. In this work, we provide a comprehensive evaluation of ASR performance on a dataset containing child-adult interactions from autism diagnostic sessions, using Whisper, Wav2Vec2, HuBERT, and WavLM. We find that speech foundation models show a noticeable performance drop (15-20% absolute WER) for child speech compared to adult speech in the conversational setting. Then, we fine-tune the best-performing zero-shot model (Whisper-large) using LoRA in a low-resource setting, yielding 8% and 13% absolute WER improvements for child and adult speech, respectively. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_16135 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Evaluation of Speech Foundation Models for ASR on Child-Adult Conversations in Autism Diagnostic Sessions Ashvin, Aditya Lahiri, Rimita Kommineni, Aditya Bishop, Somer Lord, Catherine Kadiri, Sudarsana Reddy Narayanan, Shrikanth Audio and Speech Processing Machine Learning Sound Reliable transcription of child-adult conversations in clinical settings is crucial for diagnosing developmental disorders like Autism. Recent advances in deep learning and availability of large scale transcribed data has led to development of speech foundation models that have shown dramatic improvements in ASR performance. However, their performance on conversational child-adult interactions remains underexplored. In this work, we provide a comprehensive evaluation of ASR performance on a dataset containing child-adult interactions from autism diagnostic sessions, using Whisper, Wav2Vec2, HuBERT, and WavLM. We find that speech foundation models show a noticeable performance drop (15-20% absolute WER) for child speech compared to adult speech in the conversational setting. Then, we fine-tune the best-performing zero-shot model (Whisper-large) using LoRA in a low-resource setting, yielding 8% and 13% absolute WER improvements for child and adult speech, respectively. |
| title | Evaluation of Speech Foundation Models for ASR on Child-Adult Conversations in Autism Diagnostic Sessions |
| topic | Audio and Speech Processing Machine Learning Sound |
| url | https://arxiv.org/abs/2409.16135 |