Knowledge Distillation for Real-Time Classification of Early Media in Voice Communications
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_ | 1866913959434518528 |
|---|---|
| author | Altwlkany, Kemal Hadžić, Hadžem Kurić, Amar Lacic, Emanuel |
| author_facet | Altwlkany, Kemal Hadžić, Hadžem Kurić, Amar Lacic, Emanuel |
| contents | This paper investigates the industrial setting of real-time classification of early media exchanged during the initialization phase of voice calls. We explore the application of state-of-the-art audio tagging models and highlight some limitations when applied to the classification of early media. While most existing approaches leverage convolutional neural networks, we propose a novel approach for low-resource requirements based on gradient-boosted trees. Our approach not only demonstrates a substantial improvement in runtime performance, but also exhibits a comparable accuracy. We show that leveraging knowledge distillation and class aggregation techniques to train a simpler and smaller model accelerates the classification of early media in voice calls. We provide a detailed analysis of the results on a proprietary and publicly available dataset, regarding accuracy and runtime performance. We additionally report a case study of the achieved performance improvements at a regional data center in India. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_21478 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Knowledge Distillation for Real-Time Classification of Early Media in Voice Communications Altwlkany, Kemal Hadžić, Hadžem Kurić, Amar Lacic, Emanuel Sound Artificial Intelligence Multimedia Audio and Speech Processing I.2.0 This paper investigates the industrial setting of real-time classification of early media exchanged during the initialization phase of voice calls. We explore the application of state-of-the-art audio tagging models and highlight some limitations when applied to the classification of early media. While most existing approaches leverage convolutional neural networks, we propose a novel approach for low-resource requirements based on gradient-boosted trees. Our approach not only demonstrates a substantial improvement in runtime performance, but also exhibits a comparable accuracy. We show that leveraging knowledge distillation and class aggregation techniques to train a simpler and smaller model accelerates the classification of early media in voice calls. We provide a detailed analysis of the results on a proprietary and publicly available dataset, regarding accuracy and runtime performance. We additionally report a case study of the achieved performance improvements at a regional data center in India. |
| title | Knowledge Distillation for Real-Time Classification of Early Media in Voice Communications |
| topic | Sound Artificial Intelligence Multimedia Audio and Speech Processing I.2.0 |
| url | https://arxiv.org/abs/2410.21478 |