Update README.md

README.md

@@ -10,12 +10,21 @@ tags:
 - block-diffusion
 - speech-synthesis
 ---
+<img width="800" alt="Resemble_AI_Chatterbox_Flash" src="https://cdn-uploads.huggingface.co/production/uploads/68af35bc5928acf3617150db/lCegiC8uo_eA3RHOVC6go.png" />
 
-# Chatterbox-Flash
+<h1 style="font-size: 32px">Chatterbox-Flash</h1>
 
-Chatterbox-Flash is a **block-diffusion zero-shot TTS model** that extends
-the [Chatterbox-TTS](https://github.com/resemble-ai/chatterbox) pipeline
-with a parallel masked decoder while preserving streaming generation.
+<div style="display: flex; align-items: center; gap: 12px">
+  <a href="https://huggingface.co/ResembleAI/chatterbox-flash">
+    <img src="https://huggingface.co/datasets/huggingface/badges/resolve/main/open-in-hf-spaces-sm.svg" alt="Open in HF" />
+  </a>
+</div>
+<div style="display: flex; align-items: center; gap: 8px;">
+  <span style="font-style: italic;white-space: pre-wrap">Made with  ❤️  by</span>
+  <img width="100" alt="resemble-logo-horizontal" src="https://github.com/user-attachments/assets/35cf756b-3506-4943-9c72-c05ddfa4e525" />
+</div>
+
+Chatterbox-Flash is a **block-diffusion zero-shot TTS model** that extends the [Chatterbox-TTS](https://github.com/resemble-ai/chatterbox) pipeline with a parallel masked decoder while preserving streaming generation.
 
 The released weights cover the four checkpoints needed at inference time:
 
@@ -26,16 +35,13 @@ The released weights cover the four checkpoints needed at inference time:
 | `ve.safetensors`            | GE2E voice encoder (taken verbatim from `ResembleAI/chatterbox`).  |
 | `tokenizer.json`            | English BPE tokenizer (taken verbatim from `ResembleAI/chatterbox`).|
 
-## Quick start
-
+# Quick start
 ```bash
 pip install chatterbox-flash
 ```
-
 ```python
 import torchaudio as ta
 from chatterbox_flash import ChatterboxFlashTTS
-
 tts = ChatterboxFlashTTS.from_pretrained("ResembleAI/chatterbox-flash", device="cuda")
 wav = tts.generate(
     "Hello, world.",
@@ -44,15 +50,13 @@ wav = tts.generate(
 ta.save("out.wav", wav.unsqueeze(0).cpu(), tts.sr)
 ```
 
-## Inference defaults (paper configuration)
-
-* Block size `D = 16`
-* Maximum `K = 10` denoising steps per block
-* Sampling temperature `0.6`
-* `shift` outlier schedule with `tau = 0.1`
-* Zero-text-batch CFG with `w = 1.5`, `pmi_cfg` combination, `zero_all` null prefix
-* FlashInfer paged KV cache + CUDA graph capture
-
-## License
+# Inference defaults (paper configuration)
+- Block size `D = 16`
+- Maximum `K = 10` denoising steps per block
+- Sampling temperature `0.2`
+- `shift` outlier schedule with `tau = 0.5`
+- CFG with `w = 1.0`, `pmi_cfg` combination
+- FlashInfer paged KV cache + CUDA graph capture
 
-MIT — see `LICENSE` in the source repository.
+# License
+MIT — see `LICENSE` in the source repository.