This paper presents a textile-based wearable human-machine interface with integrated tactile sensors and vibrotactile haptic actuators, fabricated using a digital embroidery machine. The system enables the recording, reproduction, and adaptive transfer of tactile interactions. The tactile sensors and haptic actuators are digitally designed and rapidly fabricated, allowing for customizable, scalable, and modular integration into textiles. The system includes a machine-learning pipeline that adaptively models individual user responses to haptic feedback, optimizing parameters for effective tactile interaction transfer. The interface demonstrates three end-to-end systems: alleviating tactile occlusion by transferring forces sensed on the outside of the glove to haptic sensations on the person's hand inside the glove, guiding people to perform physical skills, and enabling responsive robot teleoperation. The system uses a combination of tactile sensing and vibrotactile haptic feedback to enable physical interaction transfer across people and between humans and robots. The tactile sensing array and vibrotactile haptic array are digitally designed and fabricated, seamlessly integrated into smart gloves. Experiments demonstrate the system's ability to record, reproduce, and adaptively transfer physical interactions in various contexts. The system's performance is evaluated through user studies, showing that users can identify haptic feedback with high accuracy and that the system can effectively convey spatial and temporal information. The system also includes an adaptive human model learning and inverse haptics optimization pipeline, which allows for personalized haptic instructions based on individual user responses. This pipeline enables the system to adaptively optimize haptic feedback for different users and tasks, improving the transfer of tactile information without the need for manual calibration. The system has been tested in various applications, including piano playing, rhythmic gaming, and racing games, demonstrating its effectiveness in adaptive tactile interaction transfer across users. The system's ability to transfer tactile information between humans and robots has been demonstrated in teleoperation scenarios, where it improves the grasping of fragile and soft objects. The system's potential applications extend beyond the lab, including use in training programs for various skilled professions such as surgeons, pilots, and engineers. The textile-based wearable human-machine interface could also be integrated into virtual reality and augmented reality systems, providing users with a more immersive and tactile experience. The system's ability to adaptively transfer tactile information between humans and machines could be used in industries such as manufacturing, where robots and humans can work collaboratively with increased safety and efficiency.This paper presents a textile-based wearable human-machine interface with integrated tactile sensors and vibrotactile haptic actuators, fabricated using a digital embroidery machine. The system enables the recording, reproduction, and adaptive transfer of tactile interactions. The tactile sensors and haptic actuators are digitally designed and rapidly fabricated, allowing for customizable, scalable, and modular integration into textiles. The system includes a machine-learning pipeline that adaptively models individual user responses to haptic feedback, optimizing parameters for effective tactile interaction transfer. The interface demonstrates three end-to-end systems: alleviating tactile occlusion by transferring forces sensed on the outside of the glove to haptic sensations on the person's hand inside the glove, guiding people to perform physical skills, and enabling responsive robot teleoperation. The system uses a combination of tactile sensing and vibrotactile haptic feedback to enable physical interaction transfer across people and between humans and robots. The tactile sensing array and vibrotactile haptic array are digitally designed and fabricated, seamlessly integrated into smart gloves. Experiments demonstrate the system's ability to record, reproduce, and adaptively transfer physical interactions in various contexts. The system's performance is evaluated through user studies, showing that users can identify haptic feedback with high accuracy and that the system can effectively convey spatial and temporal information. The system also includes an adaptive human model learning and inverse haptics optimization pipeline, which allows for personalized haptic instructions based on individual user responses. This pipeline enables the system to adaptively optimize haptic feedback for different users and tasks, improving the transfer of tactile information without the need for manual calibration. The system has been tested in various applications, including piano playing, rhythmic gaming, and racing games, demonstrating its effectiveness in adaptive tactile interaction transfer across users. The system's ability to transfer tactile information between humans and robots has been demonstrated in teleoperation scenarios, where it improves the grasping of fragile and soft objects. The system's potential applications extend beyond the lab, including use in training programs for various skilled professions such as surgeons, pilots, and engineers. The textile-based wearable human-machine interface could also be integrated into virtual reality and augmented reality systems, providing users with a more immersive and tactile experience. The system's ability to adaptively transfer tactile information between humans and machines could be used in industries such as manufacturing, where robots and humans can work collaboratively with increased safety and efficiency.