Update README.md

README.md

@@ -134,7 +134,7 @@ Evaluations can be very long (especially for whisper) so we selected only segmen
 
 ![WER table](https://huggingface.co/linagora/linto_stt_fr_fastconformer/resolve/main/assets/wer_table.png)
 
-As shown in the table above, the model demonstrates robust performance across all datasets, consistently achieving results close to the best.
+As shown in the table above (lower is better), the model demonstrates robust performance across all datasets, consistently achieving results close to the best.
 
 ### Real-Time Factor (RTF)
 
@@ -143,6 +143,7 @@ RTFX (the inverse of RTF) measures how many seconds of audio can be transcribed
 Evaluation:
 - Hardware: Laptop with NVIDIA RTX 4090
 - Input: 5 audio files (~2 minutes each) from the ACSYNT corpus
+- Higher is better
 
 ![RTF table](https://huggingface.co/linagora/linto_stt_fr_fastconformer/resolve/main/assets/rtf_table.png)