Consolidate Results section to single summary chart

README.md

@@ -63,23 +63,11 @@ Automatic speech recognition systems are expected to faithfully transcribe speec
 
 ## Results
 
-MiMo-V2.5-ASR has been evaluated across a broad set of benchmarks spanning standard Mandarin and English, Chinese dialects, singing, code-switching, noisy conditions, and multi-speaker scenarios. Highlights of our results are shown below.
+MiMo-V2.5-ASR has been evaluated across a broad set of benchmarks spanning standard Mandarin and English, Chinese dialects, lyric recognition, and internal business scenarios. The chart below summarizes the average performance of MiMo-V2.5-ASR across these scenarios.
 
-### Standard Mandarin Chinese
+![Results](https://github.com/XiaomiMiMo/MiMo-V2.5-ASR/raw/main/assets/results_overall.png)
 
-![Results - Standard Chinese](https://github.com/XiaomiMiMo/MiMo-V2.5-ASR/raw/main/assets/results_chinese.png)
-
-### Standard English
-
-![Results - Standard English](https://github.com/XiaomiMiMo/MiMo-V2.5-ASR/raw/main/assets/results_english.png)
-
-### Chinese Dialects
-
-![Results - Chinese Dialects](https://github.com/XiaomiMiMo/MiMo-V2.5-ASR/raw/main/assets/results_dialect.png)
-
-### Singing & Code-Switch
-
-![Results - Singing and Code-Switch](https://github.com/XiaomiMiMo/MiMo-V2.5-ASR/raw/main/assets/results_singing_codeswitch.png)
+For per-benchmark numbers and specific qualitative cases, please refer to our [blog](https://xiaomimimo.github.io/MiMo-V2.5-ASR-Demo).
 
 ## Model Download