Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
|
@@ -176,14 +176,14 @@ def process_audio_rag(audio_file_path, query, chunk_length=30, use_openai=False,
|
|
| 176 |
# Create Gradio interface
|
| 177 |
with gr.Blocks(title="AudioRAG Demo") as demo:
|
| 178 |
gr.Markdown("# AudioRAG Demo - Semantic Audio Search")
|
| 179 |
-
|
| 180 |
-
|
| 181 |
|
| 182 |
-
|
| 183 |
-
|
| 184 |
|
| 185 |
-
|
| 186 |
-
|
| 187 |
|
| 188 |
with gr.Row():
|
| 189 |
with gr.Column():
|
|
|
|
| 176 |
# Create Gradio interface
|
| 177 |
with gr.Blocks(title="AudioRAG Demo") as demo:
|
| 178 |
gr.Markdown("# AudioRAG Demo - Semantic Audio Search")
|
| 179 |
+
gr.Markdown("""
|
| 180 |
+
This demo builds on the work from the ColQwen team, expanding retrieval capabilities beyond images to include audio and video. Inspired by the Qwen-Omni series, ColQwen-Omni (3B) pushes the boundaries of multimodal search — embedding and retrieving almost any type of content.
|
| 181 |
|
| 182 |
+
**What’s new?**
|
| 183 |
+
Unlike traditional methods, this model searches directly through raw audio without converting it to text. It understands semantic meaning in sound, speech, and audio patterns — making "AudioRAG" a real possibility.
|
| 184 |
|
| 185 |
+
📖 [Blog post](https://huggingface.co/blog/manu/colqwen-omni-omnimodal-retrieval) | 🤗 [Model on Hugging Face](https://huggingface.co/vidore/colqwen-omni-v0.1)
|
| 186 |
+
""")
|
| 187 |
|
| 188 |
with gr.Row():
|
| 189 |
with gr.Column():
|