Hear through (HT) filter is designed to provide a listening experience similar to open ear listening. The key challenge in the design of the HT filter is to ensure low processing delay to closely match the open ear response. Existing adaptive HT approaches such as adaptive finite impulse response (FIR) and neural network-based approaches can lead to higher processing delays. This paper proposes a sub-band based infinite impulse response (IIR) adaptive feedforward equalization method to improve the HT and convergence performance for sound sources incident from different directions. To ensure IIR filter stability, we propose inclusion of stability constraint based on least mean square fourth criterion. The proposed method improves performance over existing methods up to 13 dB for simulated outdoor and indoor scenarios, while maintaining stability and similar complexity as existing adaptive IIR and FIR techniques in simulated complex indoor and outdoor real-world acoustic environments.