Update examples to show code-switching improvements

- Replaced examples with cleaner code-switching demonstrations
- All examples now show both Chinese and English content
- Removed preamble-related example
- Examples show: translation prevention, technical term preservation, complex code-switching

README.md

@@ -40,45 +40,45 @@ A LoRA adapter for [Qwen/Qwen2-Audio-7B-Instruct](https://huggingface.co/Qwen/Qw
 
 Below are examples showing improvements from baseline to DPO-trained model:
 
-### Example 1: Removed Unnecessary Translation
+### Example 1: Code-Switching Preserved (Lifestyle)
 | | Transcription |
 |---|---|
-| **Ground Truth** | how long she is been in the church |
-| **Baseline** | 她已经在这个教堂里待了五年多了。 *(Chinese translation)* |
-| **This Model** | how long she's been in the church. |
-| **MER** | 2.62 → **0.25** |
+| **Ground Truth** | 我们 都 应该 pursue a healthy lifestyle |
+| **Baseline** | 我们都应该追求健康的生活方式 *(fully translated to Chinese)* |
+| **This Model** | 我们都应该 pursue a healthy lifestyle |
+| **MER** | 1.00 → **0.00** |
 
-### Example 2: Code-Switching Preserved
+### Example 2: Mixed Language Preserved (Christmas)
 | | Transcription |
 |---|---|
-| **Ground Truth** | (err) 你 you can choose primary school or secondary school |
-| **Baseline** | 呃你你可以选择公立学校或私立学校。 *(fully translated)* |
-| **This Model** | 呃你 you can choose primary school or secondary school |
-| **MER** | 2.20 → **0.10** |
+| **Ground Truth** | every christmas 我 就 应该 是 没有 人 跟我 庆祝 了 [啦] |
+| **Baseline** | every christmas i would - should be no one to tell me *(Chinese translated to English)* |
+| **This Model** | every christmas 我就应该是没有人跟我庆祝了啦 |
+| **MER** | 0.88 → **0.00** |
 
-### Example 3: Mixed Language Preserved
+### Example 3: Technical Terms Preserved
 | | Transcription |
 |---|---|
-| **Ground Truth** | 妳 去 过 which countries |
-| **Baseline** | 你去过哪个国家 *(fully translated to Chinese)* |
-| **This Model** | 你去过 which country |
-| **MER** | 2.40 → **0.40** |
+| **Ground Truth** | (呃) 每个 lecture different lecturer 那个 notes 不 不怎么 好的 [啦] |
+| **Baseline** | 呃那个老师不同风格的老师 *(lost technical terms)* |
+| **This Model** | 呃 每个 lecture different lecturer 那个 notes 不不怎么好的啦 |
+| **MER** | 0.75 → **0.00** |
 
-### Example 4: Clean Output Without Preamble
+### Example 4: Complex Code-Switching Preserved
 | | Transcription |
 |---|---|
-| **Ground Truth** | then you suggest english |
-| **Baseline** | The original content of this audio is: 'Then you suggest English? |
-| **This Model** | Then you suggest English. |
-| **MER** | 1.75 → **0.00** |
+| **Ground Truth** | [哦] 还有 什么 好吃 的 吗 还是 你 只是 去 那些 very expensive places like dempsey to eat |
+| **Baseline** | Oh, what else? Oh, yeah, there's always that expensive place like... to eat *(lost Chinese content)* |
+| **This Model** | 哦 还有什么好吃的吗 还是你只是去那些 very expensive places like dancy to eat |
+| **MER** | 0.83 → **0.04** |
 
-### Example 5: Code-Switching Session Preserved
+### Example 5: Professional Terms Preserved
 | | Transcription |
 |---|---|
-| **Ground Truth** | 你们 是 一首 歌 也是 教 一个 session [啊] [哦] [嗯] |
-| **Baseline** | 你们是一首歌也是叫一个session啊哦我们就是来拍一次session的product然后... *(hallucinated extra content)* |
-| **This Model** | 你们是一首歌也是叫一个 session 哦 |
-| **MER** | 1.80 → **0.20** |
+| **Ground Truth** | [哦] 因为 是个 professional degree |
+| **Baseline** | 哦因为他有个专业的学位 *(translated to Chinese)* |
+| **This Model** | 哦 因为 是个 professional degree |
+| **MER** | 1.00 → **0.00** |
 
 ## Training Configuration