• Vikash Kumar
• Tanvi Rustagi

The global challenge of women’s safety continues to grow because harassment incidents and urgent emergency situations both increase. Users who want safety today must activate emergency protocols by pressing buttons or by using GPS. Some real-world emergency situations do not respond well to emergency response procedures that are already in use. Users can activate emergency functions automatically via AI Voice recognition and fuzzy keyword search features in the 'Friend in Threat' application. The research develops a smart safety application by uniting NLP and ML capabilities with Android Framework to activate distress alerts which deliver pre-selected users location and identity details. This paper examines present safety applications before showing how AI voice-based interventions quicken response times during emergencies with a clear explanation of these enhancements. This paper explores detailed technical information about algorithms along with a presentation of current obstacles and future developments which include offline NLP processing and deep learning models for voice pattern recognition systems

Friend in Threat

"The ‘Friend in Threat’ Voice Operated AI-System ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 5, page no.b539-b548, May-2025, Available :http://www.jetir.org/papers/JETIR2505160.pdf

"The ‘Friend in Threat’ Voice Operated AI-System ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 5, page no. ppb539-b548, May-2025, Available at : http://www.jetir.org/papers/JETIR2505160.pdf