<![CDATA[ABSTRAKPusat Penelitian Teh dan Kina (PPTK) Gambung menghadapi tantangan dalam memberikan informasi cepat kepada pengunjung. Untuk mengatasinya, dikembangkan voicebot interaktif berbasis teknologi Python Speech Recognition dan Pyttsx3, yang menggunakan metode speech-to-text dan text-to-speech. Pengujian dilakukan dengan variasi kondisi internet, intensitas kebisingan, perbedaan aksen, dan analisis keluaran suara. Hasil menunjukkan akurasi hingga 90% dengan rata-rata kecepatan tanggap 1,94 detik pada koneksi internet stabil dan suara yang jelas. Di lingkungan bising dengan kekuatan suara tinggi (105 dB), voicebot tetap mampu menanggapi. Voicebot ini juga menunjukkan akurasi yang sama (90%) untuk penutur asli dan non-asli bahasa Inggris. Solusi ini berpotensi meningkatkan aksesibilitas informasi di PPTK Gambung, meskipun kinerjanya dipengaruhi oleh kestabilan internet dan kondisi lingkungan.Kata kunci: voicebot, voice-to-text, text-to-speech, pyttsx3, speech recognition ABSTRACTThe Gambung Tea and Cinchona Research Center (PPTK) faces challenges in providing timely information to visitors. To address this, an interactive voicebot based on Python Speech Recognition technology and Pyttsx3 was developed, utilizing speech-to-text and text-to-speech methods. Tests were conducted with variations in internet conditions, noise intensity, accent differences, and voice output analysis. Results show accuracy of up to 90% with an average response speed of 1,94 seconds on a stable internet connection and clear voice. In noisy environments with high voice strength (105 dB), the voicebot was still able to respond. The voicebot also showed similar accuracy (90%) for native and nonnative English speakers. This solution has the potential to improve information accessibility at PPTK Gambung, although its performance is affected by internet stability and environmental conditions.Keywords: voicebot, voice-to-text, text-to-speech, pyttsx3, speech recognition]]>
Copyrights © 2024
