Update constants.py

constants.py

@@ -1,6 +1,6 @@
 from pathlib import Path
 
-# Directory where request by models are stored
+# Directory where model requests are stored
 DIR_OUTPUT_REQUESTS = Path("requested_models")
 EVAL_REQUESTS_PATH = Path("eval_requests")
 
@@ -8,17 +8,21 @@ EVAL_REQUESTS_PATH = Path("eval_requests")
 # Text definitions       #
 ##########################
 
-banner_url = "https://huggingface.co/datasets/vargha/persian_asr_leaderboard/resolve/main/banner.png"
+banner_url = "https://cdn-thumbnails.huggingface.co/social-thumbnails/spaces/k2-fsa/automatic-speech-recognition.png"
 BANNER = f'<div style="display: flex; justify-content: space-around;"><img src="{banner_url}" alt="Banner" style="width: 40vw; min-width: 300px; max-width: 600px;"> </div>'
 
-INTRODUCTION_TEXT = "📐 The 🤗 Persian ASR Leaderboard ranks and evaluates speech recognition models \
-on the Hugging Face Hub using the Persian Common Voice dataset. \
-\nWe report the [WER](https://huggingface.co/spaces/evaluate-metric/wer) and [CER](https://huggingface.co/spaces/evaluate-metric/cer) metrics (⬇️ lower the better). Models are ranked based on their WER, from lowest to highest. Check the 📈 Metrics tab to understand how the models are evaluated. \
-\nIf you want results for a model that is not listed here, you can submit a request for it to be included ✉️✨."
+INTRODUCTION_TEXT = """
+📐 The 🤗 **Persian Automatic Speech Recognition (ASR) Leaderboard** serves as an authoritative ranking of speech recognition models hosted on the Hugging Face Hub, evaluated using multiple Persian speech datasets. 
+We report two key performance metrics: [Word Error Rate (WER)](https://huggingface.co/spaces/evaluate-metric/wer) and [Character Error Rate (CER)](https://huggingface.co/spaces/evaluate-metric/cer), where lower scores indicate better performance. 
+
+The leaderboard primarily ranks models based on WER, from lowest to highest. You can refer to the 📈 **Metrics** tab for a detailed explanation of how these models are evaluated.
+
+If there is a model you'd like to see ranked but is not listed here, you may submit a request for evaluation by following the instructions in the "Request a Model" tab ✉️✨.
+"""
 
 CITATION_TEXT = """@misc{persian-asr-leaderboard,
     title        = {Persian Automatic Speech Recognition Leaderboard},
-    author       = {Your Name},
+    author       = {Navid},
     year         = 2024,
     publisher    = {Hugging Face},
     howpublished = "\\url{https://huggingface.co/spaces/your-username/persian_asr_leaderboard}"
@@ -26,22 +30,28 @@ CITATION_TEXT = """@misc{persian-asr-leaderboard,
 """
 
 METRICS_TAB_TEXT = """
-# Metrics and Dataset
-
+# Evaluation Metrics and Datasets
 ## Metrics
+We employ the following metrics to evaluate the performance of Automatic Speech Recognition (ASR) models:
+- **Word Error Rate (WER)**: WER quantifies the proportion of incorrectly predicted words in a transcription. A lower WER reflects higher accuracy.
+- **Character Error Rate (CER)**: CER measures errors at the character level, providing a more granular view of transcription accuracy, especially in morphologically rich languages such as Persian.
 
-We evaluate models using the Word Error Rate (WER) and Character Error Rate (CER) metrics. Both metrics are used to measure the accuracy of automatic speech recognition systems.
+Both metrics are widely used in ASR evaluation, offering a comprehensive view of model performance.
 
-- **Word Error Rate (WER)**: Calculates the percentage of words that were incorrectly predicted. A lower WER indicates better performance.
-- **Character Error Rate (CER)**: Similar to WER but operates at the character level, which can be more informative for languages with rich morphology like Persian.
+## Datasets
+The models on the Persian ASR Leaderboard are evaluated using a diverse range of datasets to ensure robust performance across different speech conditions:
 
-## Dataset
+1. **Persian Common Voice (Mozilla)**  
+   Available [here](https://huggingface.co/datasets/mozilla-foundation/common_voice_17_0), this dataset is part of the broader Common Voice project and features speech data from various speakers, accents, and environments. It serves as a representative benchmark for Persian ASR.
 
-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.
+2. **ASR Farsi YouTube Chunked 10 Seconds**  
+   This dataset, available on Hugging Face [here](https://huggingface.co/datasets/pourmand1376/asr-farsi-youtube-chunked-10-seconds), consists of transcribed speech from Persian YouTube videos, split into 10-second segments. It introduces variability in audio quality and speaker demographics, adding to the challenge of accurate recognition.
 
-## How to Submit Your Model
+3. **Persian-ASR-Test-Set (Private)**  
+   This private dataset is designed for in-depth model testing and evaluation. It contains curated, real-world Persian speech data from various contexts and speaker backgrounds. Access to this dataset is restricted, ensuring models are evaluated on a controlled, high-quality speech corpus.
 
-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.
+## How to Submit Your Model for Evaluation
+To request that a model be included on this leaderboard, please submit its name in the following format: `username/model_name`. Models should be available on the Hugging Face Hub for public access.
 
+Simply navigate to the "Request a Model" tab, enter the details, and your model will be evaluated at the next available opportunity.
 """
-