CNN achieves 91.79% accuracy for Hindi keyword spotting in speech recognition

Researchers have developed a keyword spotting system for Hindi speech recognition using a Convolutional Neural Network (CNN). The system was trained on 40,000 audio samples and utilizes Mel Frequency Cepstral Coefficients (MFCCs) as input for the CNN. Experiments with various CNN architectures demonstrated a notable accuracy of 91.79% for identifying keywords in continuous Hindi speech, emphasizing computational efficiency and user-specific customization. AI

影响 Introduces a novel CNN-based approach for Hindi keyword spotting, potentially improving on-device voice command accuracy and customization.

排序理由 This is a research paper detailing a new method for keyword spotting in Hindi speech recognition. [lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

• arXiv
• CNN
• Convolutional Neural Network
• Mel Frequency Cepstral Coefficients
• MFCCs

• 论文
• 其他

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →