Article 1 Building MoulSot: How We Curated 1,500 Hours of Moroccan Darija Speech, Selected the Best 80 Hours for Transcription, and Fine-Tuned Qwen3-ASR on Top of It
MoulSot.0.2 Series of Moroccan Darija Speech Systems and Datasets. v0.2 01Yassine/moulsot_v0.2_16000 Automatic Speech Recognition • 0.8B • Updated Oct 5, 2025 • 4 01Yassine/moulsot_v0.2_35000 2B • Updated Dec 24, 2025 • 69 01Yassine/moulsot_v0.2_12500 0.8B • Updated Dec 24, 2025 • 135
MoulSot.0.1 Series of Moroccan Darija Speech Systems and Datasets 01Yassine/moulsot_v0.1_18000 Automatic Speech Recognition • 0.8B • Updated Oct 4, 2025 • 7 01Yassine/moulsot_v0.2_16000 Automatic Speech Recognition • 0.8B • Updated Oct 5, 2025 • 4 01Yassine/moulsot_v0.1_2500 Automatic Speech Recognition • 0.8B • Updated Oct 4, 2025 • 6 01Yassine/moulsot_v0.2_1000 Automatic Speech Recognition • 0.8B • Updated Oct 5, 2025 • 4
MoulSot.0.2 Series of Moroccan Darija Speech Systems and Datasets. v0.2 01Yassine/moulsot_v0.2_16000 Automatic Speech Recognition • 0.8B • Updated Oct 5, 2025 • 4 01Yassine/moulsot_v0.2_35000 2B • Updated Dec 24, 2025 • 69 01Yassine/moulsot_v0.2_12500 0.8B • Updated Dec 24, 2025 • 135
MoulSot.0.1 Series of Moroccan Darija Speech Systems and Datasets 01Yassine/moulsot_v0.1_18000 Automatic Speech Recognition • 0.8B • Updated Oct 4, 2025 • 7 01Yassine/moulsot_v0.2_16000 Automatic Speech Recognition • 0.8B • Updated Oct 5, 2025 • 4 01Yassine/moulsot_v0.1_2500 Automatic Speech Recognition • 0.8B • Updated Oct 4, 2025 • 6 01Yassine/moulsot_v0.2_1000 Automatic Speech Recognition • 0.8B • Updated Oct 5, 2025 • 4