Spoken language identification on 4 Indonesian local languages using deep learning

Language identification is at the forefront of assistance in many applications, including multilingual speech systems, spoken language translation, recognition, and human-machine interaction via voice. The indonesian local languages using technology has enormous potential to advance tourism digital content Indonesia. goal this study identify four Indonesian languages: Javanese, Sundanese, Minangkabau, Buginese, utilizing deep learning classification techniques such as artificial neural network (ANN), convolutional (CNN), long-term short memory (LSTM). selected extraction feature for audio data employs mel-frequency cepstral coefficient (MFCC). results showed that LSTM model had highest accuracy each duration (3 s, 10 30 s), followed by CNN ANN models.