Logo Lanfrica

Financial Inclusion Speech Dataset

Domaine:

natural language processing

Type de record:

dataset
Train Automatic Speech Recognition (ASR) models for Ghanaian languages in financial contexts - Develop voice-based financial chatbots and voice-USSD interfaces for mobile money services - Build Text-to-Speech (TTS) systems to deliver financial information via voice - Create language models for financial NLP in low-resource African languages - Research on digital financial inclusion barriers and language use in underserved communities Notes / challenges: Dataset size is likely modest (tens of hours) — may require augmentation for robust ASR - Transcriptions may vary in quality or standardization