README.md: reporting correct number after finding bug in inference ti…

…me code

README.md

@@ -12,7 +12,7 @@ using only a single network, trained using only a single cost function:
 maximizing the likelihood of the training data, which makes the training
 procedure simple and stable.
 
-Our [PyTorch] implementation produces audio samples at a rate of 4850
+Our [PyTorch] implementation produces audio samples at a rate of 1200 
 kHz on an NVIDIA V100 GPU. Mean Opinion Scores show that it delivers audio
 quality as good as the best publicly available WaveNet implementation.
 
@@ -88,4 +88,4 @@ with fused residual and skip connections.
 [published model]: https://drive.google.com/open?id=1rpK8CzAAirq9sWZhe9nlfvxMF1dRgFbF
 [mel-spectrograms]: https://drive.google.com/file/d/1g_VXK2lpP9J25dQFhQwx7doWl_p20fXA/view?usp=sharing
 [LJ Speech Data]: https://keithito.com/LJ-Speech-Dataset
-[Apex]: https://github.com/nvidia/apex
+[Apex]: https://github.com/nvidia/apex