regak/Spoken-Swahili-Digit-Dataset

# Spoken Swahili Digit Dataset Spoken Swahili Digit Dataset (SSDD) is a spoken digit dataset for Swahili language. This language is spoken in east and central Africa. The speech utterances were recorded from 8 regions of Tanzania. The recordings were recorded at diverse office environment conditions with different noise conditions such as: air-conditioner sounds, walking sounds, door open or close sounds, phone ringing sounds, people talking sounds, bird sounds, and car horn sounds. The quality of the recordings is at 8kHz in wav files. All recordings are trimmed and padded to a fixed length of 0.74 seconds. To the best of our knowledge, this is the first public Swahili spoken digit dataset. FSSDD will continue to grow over time as data is contributed from different parts of East African Community. ## Current status: * 30 speakers * 3000 recordings (100 per each digit per speaker) * Swahili pronunciations * 8 Regions- Dar-es-salaam, Kilimanjaro, Kagera, Njombe, Pwani, Mara, Rukwa, and Tanga. ## Naming format Files are named in the following format similar to that of FSDD : {digitLabel}_{speakerName}_{index}.wav ## Meta-data: Coming soon! (This will include extra information like Gender, Age, Regions etc) # Contributions: Coming soon! (Online instructions) # Included utilities: Coming soon! (For trimming and padding to 0.74secs or other define length) # Release Note: - For more details, kindly refer to the following paper: Coming soon! - If you use this dataset for research work please cite this paper as: **Swahili Speech Dataset Development and Improved Pre-training Method for Spoken Digit Recognition, Alexander R. Kivaisi, Qingjie Zhao, Jimmy T. Mbelwa, ACM Transactions on Asian and Low-Resource Language Information Processing, Volume 22 Issue 7**