This paper presents a method for estimating and reconstructing sound fields within a room using physics-informed neural networks (PINNs). By incorporating a limited set of experimental room impulse responses as training data, the approach combines neural network processing capabilities with the underlying physics of sound propagation, as articulated by the wave equation. The network's ability to estimate particle velocity and intensity, in addition to sound pressure, demonstrates its capacity to represent the flow of acoustic energy and completely characterize the sound field with only a few measurements. The study also investigates the potential of this network as a tool for improving acoustic simulations, offering grid-free sound field mappings with minimal inference time. Comparative analyses against current approaches for sound field reconstruction, including data-driven techniques and elementary wave-based regression methods, are conducted. The results show that the physics-informed neural network stands out in reconstructing the early part of the room impulse response while allowing for complete sound field characterization in the time domain. The paper highlights the advantages of PINNs in both simulated and real-world acoustic settings, making them a promising tool for advancing sound field reconstruction, analysis, and reproduction methods.This paper presents a method for estimating and reconstructing sound fields within a room using physics-informed neural networks (PINNs). By incorporating a limited set of experimental room impulse responses as training data, the approach combines neural network processing capabilities with the underlying physics of sound propagation, as articulated by the wave equation. The network's ability to estimate particle velocity and intensity, in addition to sound pressure, demonstrates its capacity to represent the flow of acoustic energy and completely characterize the sound field with only a few measurements. The study also investigates the potential of this network as a tool for improving acoustic simulations, offering grid-free sound field mappings with minimal inference time. Comparative analyses against current approaches for sound field reconstruction, including data-driven techniques and elementary wave-based regression methods, are conducted. The results show that the physics-informed neural network stands out in reconstructing the early part of the room impulse response while allowing for complete sound field characterization in the time domain. The paper highlights the advantages of PINNs in both simulated and real-world acoustic settings, making them a promising tool for advancing sound field reconstruction, analysis, and reproduction methods.