Upload README.md with huggingface_hub

README.md

@@ -21,83 +21,109 @@ tags:
 - tts
 - benchmark
 - indian-languages
-dataset_info:
-  features:
-  - name: text
-    dtype: string
-  - name: language
-    dtype: string
-  - name: usecase
-    dtype: string
-  - name: eval_category
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 801182
-    num_examples: 1815
-  download_size: 355512
-  dataset_size: 801182
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
+- telephony
+- evaluation
+- multilingual
 ---
 
 # TTS General Benchmark
 
-A multilingual Text-to-Speech (TTS) benchmark dataset covering 11 Indian languages across 13 diverse use cases.
+A multilingual Text-to-Speech (TTS) evaluation benchmark covering 11 Indian languages across multiple real-world use cases. The dataset is designed for systematic and repeatable evaluation of TTS systems under both high-quality and telephony bandwidth conditions.
 
-## Dataset Description
+**Total prompts:** 1,815 unique text samples  
+**Languages:** 11  
+**Evaluation tracks:** High Quality + 8 kHz Telephony  
 
-This dataset is designed for evaluating TTS systems on Indian languages. It contains 1,261 unique text samples spanning real-world scenarios from conversational AI to audiobook narration.
+> This is an **evaluation-only benchmark dataset** intended for testing and comparison — not for model training.
 
-### Languages
+---
+
+# Dataset Overview
+
+The TTS General Benchmark provides diverse prompts that reflect practical deployment scenarios such as conversational agents, announcements, narration, support calls, and telephony bots. Prompts are curated to test clarity, robustness, pronunciation handling, and expressive capability.
+
+Each prompt is labeled with:
+
+- language
+- usecase
+- eval_category (evaluation track)
+
+The dataset contains **two independently evaluated tracks** with different prompt distributions.
+
+---
 
-Bengali, English, Gujarati, Hindi, Kannada, Malayalam, Marathi, Odia, Punjabi, Tamil, Telugu
+# Evaluation Categories
 
-### Use Cases
+## high_quality
+
+Full-band prompts intended for studio / wideband TTS evaluation. These focus on naturalness, expressiveness, and content realism.
+
+### High Quality Use Cases
 
 | Use Case | Samples |
-|----------|---------|
-| Conversational Bots | 274 |
+|----------|----------|
+| Conversational Bots | 275 |
 | Audiobook | 132 |
 | Information Narration / News | 121 |
 | General Conversations | 110 |
 | Education | 110 |
-| AI Assistants | 109 |
-| Content Creation | 109 |
-| Culture | 76 |
-| Announcements | 66 |
-| Emergency Alert | 11 |
-| Public Notice | 33 |
+| AI Assistants | 110 |
+| Content Creation | 110 |
+| Culture | 77 |
+| Announcements | 110 |
 | Indianisms | 55 |
 | Insane Repetition | 55 |
 
-## Dataset Structure
+**High-quality total:** 1,265
 
-Each sample contains the following fields:
+---
 
-```json
-{
-  "text": "The text to be synthesized",
-  "language": "hi",
-  "usecase": "Conversational Bots"
-}
-```
+## 8khz_telephony
+
+Narrowband prompts designed for telephony and call-center evaluation (8 kHz playback target). These measure intelligibility, clarity, and robustness under bandwidth constraints.
+
+### Telephony Use Cases
+
+| Use Case | Samples |
+|----------|----------|
+| collections | 110 |
+| edge_cases | 110 |
+| sales_bot | 110 |
+| support | 110 |
+| survey_bot | 110 |
+
+**Telephony total:** 550
+
+---
 
-### Fields
+# Supported Languages
 
-- **text** (`string`): The input text for TTS synthesis
-- **language** (`string`): ISO 639-1 language code
-- **usecase** (`string`): The application domain/category of the text
+| Language | Code |
+|-----------|--------|
+English | en |
+Hindi | hi |
+Bengali | bn |
+Tamil | ta |
+Telugu | te |
+Kannada | kn |
+Malayalam | ml |
+Marathi | mr |
+Gujarati | gu |
+Odia | od |
+Punjabi | pa |
 
-## Usage
+Language coverage is shared across both evaluation tracks.
 
-```python
-from datasets import load_dataset
+---
 
-dataset = load_dataset("sarvamai/tts-general-benchmark")
-```
+# Dataset Structure
 
+Each JSONL row contains:
 
+```json
+{
+  "text": "The text to be synthesized",
+  "language": "hi",
+  "usecase": "Conversational Bots",
+  "eval_category": "high_quality"
+}