Update README.md

README.md

@@ -66,22 +66,18 @@ pipeline_tag: text-to-speech
 </div>
 
 > [!IMPORTANT]
-> **Important Sampling Considerations Across Different Backends**  
->
-> **OuteTTS Version 1.0** supports multiple backends; however, since each handles sampling differently,
-> **llama.cpp** delivers the most reliable and consistent output quality by default.
-> For optimal results, I recommend using the **llama.cpp** backend with this model.
->
-> I also strongly recommend using the model with the specified settings [here](#sampling-configuration).
-> Deviating from these settings may result in low quality or broken outputs.
-> This issue stems primarily from how different backends implement the **repetition penalty**.
-> This model performs best with a **windowed approach** (using a **64-token window**), where the penalty is applied only to the most recent tokens, rather than across the entire context window.
->
-> **Llama.cpp** and **EXL2** support such sampling, while **Transformers** don't.
-> To address this, I've implemented a **windowed repetition penalty** for the **Hugging Face Transformers** backend in the **OuteTTS** library, which significantly improves output quality and resolves sampling issues, providing comparable results to llama.cpp.
-> Without this adjustment, output quality may suffer considerably.
->
-> If you've found alternative sampling settings that improve performance, please share your findings by opening an issue on the [OuteTTS GitHub](https://github.com/edwko/OuteTTS/issues).
+> **Important Sampling Considerations**  
+> 
+> When using OuteTTS version 1.0, it is crucial to use the settings specified in the [Sampling Configuration](#sampling-configuration) section.
+> 
+> The **repetition penalty implementation** is particularly important - this model requires penalization applied to a **64-token recent window**,
+> rather than across the entire context window. Penalizing the entire context will cause the model to produce **broken or low-quality output**.
+> 
+> Currently, **llama.cpp** delivers the most reliable and consistent output quality by default.
+> Both **llama.cpp** and **EXL2** support this windowed sampling approach, while **Transformers** doesn't.
+> 
+> To address this limitation, I've implemented a **windowed repetition penalty** for the **Hugging Face Transformers** backend in the **OuteTTS** library,
+> which significantly improves output quality and resolves sampling issues, providing comparable results to llama.cpp.
 
 # OuteTTS Version 1.0
 
@@ -211,7 +207,7 @@ For optimal results with this TTS model, use the following sampling settings.
 |-------------------|----------|
 | Temperature       | 0.4      |
 | Repetition Penalty| 1.1      |
-| Repetition Range  | 64       |
+| **Repetition Range**  | **64**       |
 | Top-k             | 40       |
 | Top-p             | 0.9      |
 | Min-p             | 0.05     |