iPhoneme：基于ALS的脑-文本通信系统使用ConformerXL解码
iPhoneme: Brain-to-Text Communication for ALS Using ConformerXL Decoding
AI总结：
iPhoneme通过集成建模和交互设计，解决ALS患者脑机接口中的解码准确性和输入接口限制问题，实现高精度的脑-文本通信。
摘要：
脑机接口（BCIs）在语音恢复中的应用具有变革潜力，可帮助全球约173,000至232,500名ALS相关构音障碍患者。尽管近年来取得进展，但高性能语音BCIs仅在22至31名患者中得到验证，主要受限于神经解码准确性和实际输入接口的限制。本文提出iPhoneme，一种联合解决这些挑战的脑-文本通信系统。该系统结合基于改进Conformer架构（ConformerXL，192.9M参数）的深度学习音素解码器，以及辅助音素输入的凝视系统，以缓解眼动追踪系统的Midas touch问题。声学模型包含具有多尺度稀疏卷积和双向GRU的时序预网络，用于神经抖动校正、时间子采样以提高CTC稳定性，以及跨12个编码器块的Pre-RMSNorm稳定化，通过AdamW和余弦调度训练。在交互方面，iPhoneme引入了一种基于和弦的凝视加静默语音范式，取代停留时间选择，以提高输入效率。该系统在T15数据集（45个会话，8,071次试验）上进行评估，该数据集来自256通道的颅内EEG，来自语音运动皮层区域。一个基于310万序列训练的6元音素语言模型，结合WFST束搜索（束宽=128），实现了92.14%的音素准确率（7.86%的PER）和73.39%的词准确率（26.61%的WER），比先前的最先进水平高约3%。该系统在CPU上运行，延迟为180毫秒，展示了实时、高精度的ALS脑-文本通信。
英文摘要：
Brain-computer interfaces (BCIs) for speech restoration hold transformative potential for the approximately 173,000--232,500 individuals worldwide with ALS-related dysarthria. Despite recent progress, high-performance speech BCIs have been demonstrated in only 22--31 patients globally, largely due to limitations in neural decoding accuracy and practical input interfaces. We present iPhoneme, a brain-to-text communication system that jointly addresses these challenges through integrated modeling and interaction design. The system combines a deep learning phoneme decoder based on a modified Conformer architecture (ConformerXL, 192.9M parameters) with a gaze-assisted phoneme input interface that mitigates the Midas touch problem in eye-tracking systems. The acoustic model incorporates a temporal prenet with multi-scale dilated convolutions and bidirectional GRU for neural jitter correction, temporal subsampling for CTC stability, and Pre-RMSNorm stabilization across 12 encoder blocks, trained with AdamW and cosine scheduling. On the interaction side, iPhoneme introduces a chorded gaze-plus-silent-speech paradigm that replaces dwell-time selection, enabling more efficient input. We evaluate the system on the T15 dataset (45 sessions, 8,071 trials) of 256-channel intracranial EEG from speech motor cortex regions. A 6-gram phoneme language model trained on 3.1M sequences, combined with WFST beam search (beam=128), achieves 92.14% phoneme accuracy (7.86% PER) and 73.39% word accuracy (26.61% WER), approximately 3% above prior state-of-the-art. The system operates on CPU with 180 ms latency, demonstrating real-time, high-accuracy brain-to-text communication for ALS.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
