Researchers have developed a brain-computer interface (BCI) capable of nearly instantaneous voice synthesis from neural signals with natural speech features like intonation and singing. Implanted in a man with ALS and severe speech impairment, the system decodes neural activity from hundreds of microelectrodes in the speech motor cortex, enabling expressive communication in real time with millisecond latency. This breakthrough surpasses earlier BCIs that were slower or less natural, marking a major advance toward restoring speech for people with neurological conditions.