Update README.md

README.md

@@ -0,0 +1,78 @@
+---
+language:
+- en
+pipeline_tag: audio-text-to-text
+tags:
+- audio
+- speech
+- voice-assistant
+- voicebench
+- gemma
+---
+
+
+![LFG-3-hero](https://cdn-uploads.huggingface.co/production/uploads/61b37e66986f43ddf4956d21/EeQbXGjXA4z82M6VlegVf.png)
+
+LFG-3 is an audio-language model that fuses the intelligence of Gemma 4 31B with a Parakeet audio encoder through a trained projection later. Speech goes in, text comes out.
+
+The model is the third iteration in a personal learning journey to asnwer the question, "Can I stand on the sholders of giants and use limited compute resourcs to a standout model that can understand what and how you say things, not just speech to text." 
+
+
+## Voice Bench Results
+| Subset |Metric | 	Score	| 
+| --- | --- | ---: |
+| AlpacaEval | (1-5, GPT) |	4.73	|
+| CommonEval | (1-5, GPT) |	4.40	|
+| WildVoice | (1-5, GPT)	| 4.45	|
+| SD-QA | (% GPT majority) | 78.12	|
+| MMSU | (% accuracy)	| 85.52	8 |
+| OpenBookQA | (% accuracy)	| 94.73	|
+| BBH | (% accuracy)	| 92.20	|
+| IFEval  |(% strict-loose avg)	| 88.54	|
+|AdvBench | (% refusal rate)	| 98.27	|
+| Overall | (mean of 9)	| 89.88 |
+
+
+
+## Usage
+
+```python
+import soundfile as sf
+from transformers import AutoModelForMultimodalLM, AutoProcessor
+
+MODEL = "glenn2/LFG-3-p8-2000"
+processor = AutoProcessor.from_pretrained(MODEL, trust_remote_code=True)
+model = AutoModelForMultimodalLM.from_pretrained(
+    MODEL, trust_remote_code=True, dtype="bfloat16", device_map="cuda"
+)
+audio, sr = sf.read("question.wav")  # 16 kHz mono
+
+
+messages = [
+    {"role": "system", "content": [{"type": "text", "text": "You are a helpful voice assistant. The user is speaking to you, and your reply will be read aloud."}]},
+    {"role": "user", "content": [{"type": "audio", "audio": audio}]}, # Audio should be 16 kHz mono. 
+]
+# Process input
+inputs = processor.apply_chat_template(
+    messages,
+    tokenize=True,
+    return_dict=True,
+    return_tensors="pt",
+    add_generation_prompt=True,
+    enable_thinking=True
+).to(model.device)
+input_len = inputs["input_ids"].shape[-1]
+
+# Generate output
+outputs = model.generate(**inputs, max_new_tokens=4096)
+response = processor.decode(outputs[0][input_len:], skip_special_tokens=False)
+
+# Parse output
+print(processor.parse_response(response)["content"])  
+```
+
+
+## Intended use
+
+- Designed for **English** spoken questions/instructions → text answers.
+- Inherits knowledge from the Gemma 4 31B IT model.