Update README.md
Browse files
README.md
CHANGED
|
@@ -268,27 +268,29 @@ Here is the table based on the provided data:
|
|
| 268 |
|
| 269 |
Indic Parler-TTS has been evaluated using a MOS-like framework by native and non-native speakers. The results highlight its exceptional performance in generating natural and intelligible speech, especially for native speakers of Indian languages.
|
| 270 |
|
| 271 |
-
|
| 272 |
-
|
| 273 |
-
|
|
| 274 |
-
|
| 275 |
-
|
|
| 276 |
-
|
|
| 277 |
-
|
|
| 278 |
-
|
|
| 279 |
-
|
|
| 280 |
-
|
|
| 281 |
-
|
|
| 282 |
-
|
|
| 283 |
-
|
|
| 284 |
-
|
|
| 285 |
-
|
|
| 286 |
-
|
|
| 287 |
-
|
|
| 288 |
-
|
|
| 289 |
-
|
|
| 290 |
-
|
|
| 291 |
-
|
|
|
|
|
|
|
|
| 292 |
|
| 293 |
**Key Strengths**:
|
| 294 |
- Exceptional performance for native speakers, with top scores for **Maithili (95.36)**, **Sanskrit (99.79)**, and **Bodo (94.47)**.
|
|
|
|
| 268 |
|
| 269 |
Indic Parler-TTS has been evaluated using a MOS-like framework by native and non-native speakers. The results highlight its exceptional performance in generating natural and intelligible speech, especially for native speakers of Indian languages.
|
| 270 |
|
| 271 |
+
**NSS** stands for **Native Speaker Score**:
|
| 272 |
+
|
| 273 |
+
| **Language** | **NSS Pretrained (%)** | **NSS Finetuned (%)** | **Highlights** |
|
| 274 |
+
|----------------|-------------------------|------------------------|--------------------------------------------------------------------------------------------------|
|
| 275 |
+
| Assamese | 82.56 ± 1.80 | 87.36 ± 1.81 | Clear, natural synthesis with excellent expressiveness. |
|
| 276 |
+
| Bengali | 77.41 ± 2.14 | 86.16 ± 1.85 | High-quality outputs with smooth intonation. |
|
| 277 |
+
| Bodo | 90.83 ± 4.54 | 94.47 ± 4.12 | Near-perfect accuracy for a lesser-resourced language. |
|
| 278 |
+
| Dogri | 82.61 ± 4.98 | 88.80 ± 3.57 | Robust and consistent synthesis for Dogri. |
|
| 279 |
+
| Gujarati | 75.28 ± 1.94 | 75.36 ± 1.78 | Strong clarity and naturalness even for smaller languages. |
|
| 280 |
+
| Hindi | 83.43 ± 1.53 | 84.79 ± 2.09 | Reliable and expressive outputs for India's most widely spoken language. |
|
| 281 |
+
| Kannada | 77.97 ± 3.43 | 88.17 ± 2.81 | Highly natural and accurate voices for Kannada. |
|
| 282 |
+
| Konkani | 87.20 ± 3.58 | 76.60 ± 4.14 | Produces clear and natural outputs for diverse speakers. |
|
| 283 |
+
| Maithili | 89.07 ± 4.47 | 95.36 ± 2.52 | Exceptionally accurate, showcasing fine-tuning success. |
|
| 284 |
+
| Malayalam | 82.02 ± 2.06 | 86.54 ± 1.67 | Smooth, high-quality synthesis with expressive outputs. |
|
| 285 |
+
| Manipuri | 89.58 ± 1.33 | 85.63 ± 2.60 | Natural intonation with minimal errors. |
|
| 286 |
+
| Marathi | 73.81 ± 1.93 | 76.96 ± 1.45 | Maintains clarity and naturalness across speakers. |
|
| 287 |
+
| Nepali | 64.05 ± 8.33 | 80.02 ± 5.75 | Strong synthesis for native and proximal Nepali speakers. |
|
| 288 |
+
| Odia | 90.28 ± 2.52 | 88.94 ± 3.26 | High expressiveness and quality for Odia speakers. |
|
| 289 |
+
| Sanskrit | 99.71 ± 0.58 | 99.79 ± 0.34 | Near-perfect synthesis, ideal for classical use cases. |
|
| 290 |
+
| Sindhi | 76.44 ± 2.26 | 76.46 ± 1.29 | Clear and natural voices for underrepresented languages. |
|
| 291 |
+
| Tamil | 69.68 ± 2.73 | 75.48 ± 2.18 | Delivers intelligible and expressive speech. |
|
| 292 |
+
| Telugu | 89.77 ± 2.20 | 88.54 ± 1.86 | Smooth and natural tonal quality for Telugu. |
|
| 293 |
+
| Urdu | 77.15 ± 3.47 | 77.75 ± 3.82 | Produces high-quality speech despite resource constraints. |
|
| 294 |
|
| 295 |
**Key Strengths**:
|
| 296 |
- Exceptional performance for native speakers, with top scores for **Maithili (95.36)**, **Sanskrit (99.79)**, and **Bodo (94.47)**.
|