This is a trial run to create speech by finetuning an open source foundation model and inputting ope
A collection of read speech recordings in Kalenjin (kln).