Add Listen table: 8 lines x 2 voices (female waxal_ibo_5, male anchor), Flash-DPO

README.md

@@ -36,6 +36,8 @@ datasets:
 
 **Experimental prompt-controlled Nigerian TTS and code-switching technology** by Modular. English, Nigerian Pidgin, Igbo, Yoruba and Hausa, mixed inside one sentence and spoken by one consistent voice, with a plain-English **style instruction** (`--instruct`) that steers pace and pausing.
 
+**[Listen to samples](#listen)** (both voices, all five languages and code-switching).
+
 Pro v1.1 is a research release, not a production system. It is the Flash recipe (CosyVoice3 LM fine-tuned on licensed Nigerian-language speech, then DPO) with a new instruction layer on top. Read **What works and what does not** before using it.
 
 ## What works and what does not
@@ -121,6 +123,23 @@ Text: wrap each language span in `<eng>`, `<pcm>`, `<yor>`, `<ibo>` or `<hau>`;
 
 ## Samples
 
+### Listen
+
+The same eight lines in the two main voices, rendered with **Flash-DPO** (the default `--version flash`, default prompt): the female voice (`waxal_ibo_5`, WAXAL Igbo speaker 5) and the default male anchor. Text is shown with the language tags it was synthesised from. Mono MP3, CC-BY-SA-4.0. The Igbo and Yoruba lines have not yet been checked by a native speaker.
+
+| language | text | female (`waxal_ibo_5`) | male (anchor) |
+|---|---|---|---|
+| Igbo | &lt;ibo&gt;Ndewo, kedụ ka ị mere? Anyị ga-aga ahịa echi n'ụtụtụ.&lt;/ibo&gt; | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_01_igbo_female_waxal_ibo5.mp3"></audio> | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_01_igbo_flash_default_male.mp3"></audio> |
+| Yoruba | &lt;yor&gt;Ẹ káàárọ̀! Ṣé dáadáa ni ẹ wà? Mo fẹ́ lọ sí ọjà lọ́la.&lt;/yor&gt; | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_02_yoruba_tones_female_waxal_ibo5.mp3"></audio> | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_02_yoruba_tones_flash_default_male.mp3"></audio> |
+| Hausa | &lt;hau&gt;Sannu da zuwa. Yaya aiki? Zan tafi Kano gobe da safe.&lt;/hau&gt; | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_03_hausa_female_waxal_ibo5.mp3"></audio> | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_03_hausa_flash_default_male.mp3"></audio> |
+| Pidgin | &lt;pcm&gt;How far, my guy? Abeg make we see for evening, I dey come.&lt;/pcm&gt; | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_04_pidgin_female_waxal_ibo5.mp3"></audio> | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_04_pidgin_flash_default_male.mp3"></audio> |
+| English | &lt;eng&gt;Good morning. The meeting has moved to three o'clock this afternoon.&lt;/eng&gt; | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_05_english_female_waxal_ibo5.mp3"></audio> | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_05_english_flash_default_male.mp3"></audio> |
+| Code-switch: Igbo + English | &lt;ibo&gt;Biko,&lt;/ibo&gt; &lt;eng&gt;can you send me the report before Friday?&lt;/eng&gt; &lt;ibo&gt;Daalụ.&lt;/ibo&gt; | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_06_cs_igbo_english_female_waxal_ibo5.mp3"></audio> | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_06_cs_igbo_english_flash_default_male.mp3"></audio> |
+| Code-switch: Yoruba + English | &lt;yor&gt;Ẹ ṣé gan-an,&lt;/yor&gt; &lt;eng&gt;I will call you when I get to the office.&lt;/eng&gt; | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_07_cs_yoruba_english_female_waxal_ibo5.mp3"></audio> | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_07_cs_yoruba_english_flash_default_male.mp3"></audio> |
+| Code-switch: Pidgin + English + Hausa | &lt;pcm&gt;Na today we go finish am,&lt;/pcm&gt; &lt;eng&gt;so please be on time.&lt;/eng&gt; &lt;hau&gt;Na gode.&lt;/hau&gt; | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_08_cs_pidgin_english_hausa_female_waxal_ibo5.mp3"></audio> | <audio controls src="https://huggingface.co/Modularcomputing/dynamic-accent-pro-v1.1/resolve/main/samples/listen/tts_08_cs_pidgin_english_hausa_flash_default_male.mp3"></audio> |
+
+### All samples
+
 `samples/` holds MP3 renders for all five languages and a code-switched sentence, both main voices, with the default prompt and the slow / fast / few-pauses / many-pauses instructions, plus a Flash-DPO render of the same text and seed for A/B (`samples/meta.json` lists text, voice and instruction per file). `samples/codeswitch_frontend/` shows the experimental Qwen3-32B step: input text, request, the tagged text it produced, and the audio. The Igbo and Yoruba demo sentences have **not yet been checked by a native speaker**.
 
 ## Results