This research introduces an innovative framework for stock market crash prediction using deep learning approach by integrating quantitative financial data with qualitative sentiment derived from news headlines. The model improves predictive accuracy by integrating both quantitative and qualitative data, mirroring the growing complexity and volatility of contemporary global markets. The collection comprises quantitative financial data from prominent Vanguard sector ETFs and the VIX volatility index, along with qualitative sentiment data derived from Reddit WorldNews stories. The model architecture comprises a market data encoder, a news data encoder, and a concluding evaluation layer. The model attains an Area Under the Curve (AUC) score of roughly 0.7, underscoring the efficacy of lightweight models owing to restricted variability and temporal range of data. The primary contribution of the paper is illustrating the additional benefit of integrating qualitative news sentiment with quantitative market data to improve crash prediction.