The increasing demand for personalized, emotionally expressive synthetic voices in Text-to-Speech (TTS) systems presents the challenge of generating voices that sound genuinely friendly, especially with limited data and computational resources. This study proposes a voice cloning method for data-scarce scenarios, requiring only 10–30 minutes of speech and optimized for low-complexity environments like Google Colab. By leveraging So-VITS-SVC, a robust voice conversion model, and the acoustic precision of Parselmouth (Praat), we extract and refine vocal attributes such as intonation, pitch, and timbre for more accurate and emotionally resonant voice generation. The method was deployed in an automated answering system at HCMUT, where it generates voices that are friendly and relatable for students. Preliminary evaluations show that the system outperforms baseline models with lower F0 RMSE (19.4 Hz), higher MOS-N (4.4), MOS-I (4.7), and reduced WER (3.2%), TER (2.7%). These results highlight the effectiveness of prosody conditioning in improving voice intelligibility and naturalness. Future work will focus on refining emotional expressiveness, improving training data, and addressing ethical concerns in voice cloning.