Fine-tuned facebook/wav2vec2-xls-r-300m for speech recognition using the train split of Common Voice 7.0 (fr). When using this model, make sure that your speech input is sampled at 16kHz.

This model has been fine-tuned by the HuggingSound tool.