Table 8.

Results on the Common Voice test set with a baseline x-vector embedder.

Group	Trained on	Fine-Tuned on	Gender Results	Age Results
	TIMIT train	-	Accuracy	Weighted_F1
All			91.10%	24%
Female			75.40%	23%
Male			96.60%	24%
	Common Voice Train	-	Accuracy	Weighted_F1
All			98.00%	68%
Female			95.40%	71%
Male			98.90%	66%
	Common Voice Train	TIMIT train	Accuracy	Weighted_F1
All			94.20%	31%
Female			90.50%	28%
Male			95.50%	32%