Listening while Speaking and Visualizing: Improving ASR through Multimodal Chain
This work addresses the challenge of data scarcity in multimodal AI systems for researchers and practitioners in speech and vision domains, though it is incremental as it builds upon prior speech chain methods.
The authors tackled the problem of training automatic speech recognition (ASR) with limited paired data by constructing a multimodal chain that integrates ASR, text-to-speech synthesis, image captioning, and image production models, resulting in improved ASR performance without requiring large amounts of parallel multimodal data.
Previously, a machine speech chain, which is based on sequence-to-sequence deep learning, was proposed to mimic speech perception and production behavior. Such chains separately processed listening and speaking by automatic speech recognition (ASR) and text-to-speech synthesis (TTS) and simultaneously enabled them to teach each other in semi-supervised learning when they received unpaired data. Unfortunately, this speech chain study is limited to speech and textual modalities. In fact, natural communication is actually multimodal and involves both auditory and visual sensory systems. Although the said speech chain reduces the requirement of having a full amount of paired data, in this case we still need a large amount of unpaired data. In this research, we take a further step and construct a multimodal chain and design a closely knit chain architecture that combines ASR, TTS, image captioning, and image production models into a single framework. The framework allows the training of each component without requiring a large number of parallel multimodal data. Our experimental results also show that an ASR can be further trained without speech and text data and cross-modal data augmentation remains possible through our proposed chain, which improves the ASR performance.