Forecasting stock market movements requires exceptional complexity due to financial data displaying volatile nonlinear behavior. The research establishes a Novel Conv-LSTM Deep Learning Model for achieving advanced Indian stock market price forecasting accuracy. Stock price data undergoes spatial analysis by the Convolutional Neural Network to extract features while Long Short-Term Memory network identifies temporal patterns in the data. Both functions of the Conv-LSTM architecture unite to generate superior forecasting results by leveraging their individual strengths. A performance comparison between the proposed hybrid Conv-LSTM model and standalone CNN and LSTM models occurred using Indian stock market historical data. The Conv-LSTM model yielded better forecasting results than CNN and LSTM models through performance measurements that showed an MSE of 426.7159 and corresponding RMSE of 20.6571 and MAE of 17.7311. The errors produced by CNN were substantially higher than those from both Conv-LSTM (MSE: 1805.6726, RMSE: 42.4932, MAE: 40.5916) and LSTM (MSE: 657.3418, RMSE: 25.6387, MAE: 23.6762). The evaluation supports Conv-LSTM as an optimal selection for stock market forecasting in Indian financial markets because it detects spatial along with temporal dependencies effectively.