SW-ASR: A Context-Aware Hybrid ASR Pipeline for Robust Single Word Speech Recognition
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914288794337280 |
|---|---|
| author | Sharma, Manali Naik, Riya G, Buvaneshwari |
| author_facet | Sharma, Manali Naik, Riya G, Buvaneshwari |
| contents | Single-word Automatic Speech Recognition (ASR) is a challenging task due to the lack of linguistic context and sensitivity to noise, pronunciation variation, and channel artifacts, especially in low-resource, communication-critical domains such as healthcare and emergency response. This paper reviews recent deep learning approaches and proposes a modular framework for robust single-word detection. The system combines denoising and normalization with a hybrid ASR front end (Whisper + Vosk) and a verification layer designed to handle out-of-vocabulary words and degraded audio. The verification layer supports multiple matching strategies, including embedding similarity, edit distance, and LLM-based matching with optional contextual guidance. We evaluate the framework on the Google Speech Commands dataset and a curated real-world dataset collected from telephony and messaging platforms under bandwidth-limited conditions. Results show that while the hybrid ASR front end performs well on clean audio, the verification layer significantly improves accuracy on noisy and compressed channels. Context-guided and LLM-based matching yield the largest gains, demonstrating that lightweight verification and context mechanisms can substantially improve single-word ASR robustness without sacrificing latency required for real-time telephony applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_20890 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SW-ASR: A Context-Aware Hybrid ASR Pipeline for Robust Single Word Speech Recognition Sharma, Manali Naik, Riya G, Buvaneshwari Sound Computation and Language Audio and Speech Processing I.2.7 Single-word Automatic Speech Recognition (ASR) is a challenging task due to the lack of linguistic context and sensitivity to noise, pronunciation variation, and channel artifacts, especially in low-resource, communication-critical domains such as healthcare and emergency response. This paper reviews recent deep learning approaches and proposes a modular framework for robust single-word detection. The system combines denoising and normalization with a hybrid ASR front end (Whisper + Vosk) and a verification layer designed to handle out-of-vocabulary words and degraded audio. The verification layer supports multiple matching strategies, including embedding similarity, edit distance, and LLM-based matching with optional contextual guidance. We evaluate the framework on the Google Speech Commands dataset and a curated real-world dataset collected from telephony and messaging platforms under bandwidth-limited conditions. Results show that while the hybrid ASR front end performs well on clean audio, the verification layer significantly improves accuracy on noisy and compressed channels. Context-guided and LLM-based matching yield the largest gains, demonstrating that lightweight verification and context mechanisms can substantially improve single-word ASR robustness without sacrificing latency required for real-time telephony applications. |
| title | SW-ASR: A Context-Aware Hybrid ASR Pipeline for Robust Single Word Speech Recognition |
| topic | Sound Computation and Language Audio and Speech Processing I.2.7 |
| url | https://arxiv.org/abs/2601.20890 |