Adapting Whisper for Lightweight and Efficient Automatic Speech Recognition of Children for On-device Edge Applications
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913949587341312 |
|---|---|
| author | Dutta, Satwik Chandupatla, Shruthigna Hansen, John |
| author_facet | Dutta, Satwik Chandupatla, Shruthigna Hansen, John |
| contents | Reliability on cloud providers for ASR inference to support child-centered voice-based applications is becoming challenging due to regulatory and privacy challenges. Motivated by a privacy-preserving design, this study aims to develop a lightweight & efficient Whisper ASR system capable of running on a Raspberry Pi. Upon evaluation of the MyST corpus and by examining various filtering strategies to fine-tune the `tiny.en' model, a Word Error Rate (WER) of 15.9% was achieved (11.8% filtered). A low-rank compression reduces the encoder size by 0.51M with 1.26x faster inference in GPU, with 11% relative WER increase. During inference on Pi, the compressed version required ~2 GFLOPS fewer computations. The RTF for both the models ranged between [0.23-0.41] for various input audio durations. Analyzing the RAM usage and CPU temperature showed that the PI was capable of handling both the tiny models, however it was noticed that small models initiated additional overhead/thermal throttling. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_14451 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Adapting Whisper for Lightweight and Efficient Automatic Speech Recognition of Children for On-device Edge Applications Dutta, Satwik Chandupatla, Shruthigna Hansen, John Audio and Speech Processing Human-Computer Interaction Sound Reliability on cloud providers for ASR inference to support child-centered voice-based applications is becoming challenging due to regulatory and privacy challenges. Motivated by a privacy-preserving design, this study aims to develop a lightweight & efficient Whisper ASR system capable of running on a Raspberry Pi. Upon evaluation of the MyST corpus and by examining various filtering strategies to fine-tune the `tiny.en' model, a Word Error Rate (WER) of 15.9% was achieved (11.8% filtered). A low-rank compression reduces the encoder size by 0.51M with 1.26x faster inference in GPU, with 11% relative WER increase. During inference on Pi, the compressed version required ~2 GFLOPS fewer computations. The RTF for both the models ranged between [0.23-0.41] for various input audio durations. Analyzing the RAM usage and CPU temperature showed that the PI was capable of handling both the tiny models, however it was noticed that small models initiated additional overhead/thermal throttling. |
| title | Adapting Whisper for Lightweight and Efficient Automatic Speech Recognition of Children for On-device Edge Applications |
| topic | Audio and Speech Processing Human-Computer Interaction Sound |
| url | https://arxiv.org/abs/2507.14451 |