Update README.md
Browse files
README.md
CHANGED
|
@@ -140,4 +140,13 @@ Supervised speech instruction finetuning via knowledge-distillation. For more in
|
|
| 140 |
Check out the audio tab on [TheFastest.ai](https://thefastest.ai/?m=audio) for daily benchmarks and a comparison with other existing models.
|
| 141 |
|
| 142 |
## Evaluation
|
| 143 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 140 |
Check out the audio tab on [TheFastest.ai](https://thefastest.ai/?m=audio) for daily benchmarks and a comparison with other existing models.
|
| 141 |
|
| 142 |
## Evaluation
|
| 143 |
+
|
| 144 |
+
| | **Ultravox 0.5 1b**| Ultravox 0.5 8B | Ultravox 0.5 70B
|
| 145 |
+
| --- | ---: | ---: | ---: | --: |
|
| 146 |
+
| **covost2 en_ar** | 1.55 | 12.99 | 20.21 |
|
| 147 |
+
| **covost2 en_ca** | 8.06 | 31.54 | 40.01 |
|
| 148 |
+
| **covost2 en_de** | 14.21 | 28.70 | 34.53 |
|
| 149 |
+
| **covost2 es_en** | 24.97 | 40.19 | 43.29 |
|
| 150 |
+
| **covost2 ru_en** | 24.12 | 42.13 | 48.99 |
|
| 151 |
+
| **covost2 zh_en** | 4.76 | 17.22 | 21.37 |
|
| 152 |
+
| **big bench audio**| 39.14 | 66.54 | 82.70 |
|