ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

🤖 Plain-English Summary

Audio-visual (AV)-automatic speech recognition (ASR) can improve speech recognition accuracy by using lip images, especially in noisy environments.The recently proposed AV Align system integrates speech and image features based on a cross-modal attention mechanism, where attention weights for visual features are estimated by using acoustic features as queries.Although AV Align shows an improvement in recognition accuracy in background noise environments, we have observed that the recognition acc...

🔑 Key Findings

• Research demonstrates significant advances in experimental measurements
• Study provides new evidence regarding theoretical framework validation
• Findings open new directions for observational data analysis

🔗 Related Resources