Update constants.py

constants.py

@@ -37,14 +37,11 @@ We evaluate models using the Word Error Rate (WER) and Character Error Rate (CER
 
 ## Dataset
 
-We use the [Persian Common Voice](https://huggingface.co/datasets/vargha/common_voice_fa) dataset for evaluation. The dataset consists of diverse speech recordings from various speakers, making it a good benchmark for Persian ASR models.
+We use the [Persian Common Voice](https://huggingface.co/datasets/mozilla-foundation/common_voice_17_0) dataset for evaluation. The dataset consists of diverse speech recordings from various speakers, making it a good benchmark for Persian ASR models.
 
 ## How to Submit Your Model
 
 To submit your model for evaluation, go to the "✉️✨ Request a model here!" tab and enter your model's name in the format `username/model_name`. Your model should be hosted on the Hugging Face Hub.
 
-## Reproducing Results
-
-To reproduce the results or to see how the evaluation is conducted, you can visit our [GitHub repository](https://github.com/your-username/persian_asr_leaderboard).
 """