Update README.md

README.md

@@ -1,199 +1,193 @@
 ---
 library_name: transformers
-tags: []
+license: mit
+language:
+- multilingual
+- ar
+- zh
+- cs
+- da
+- nl
+- en
+- fi
+- fr
+- de
+- he
+- hu
+- it
+- ja
+- ko
+- no
+- pl
+- pt
+- ru
+- es
+- sv
+- th
+- tr
+- uk
+tags:
+- nlp
+- code
+- audio
+- automatic-speech-recognition
+- speech-summarization
+- speech-translation
+- phi-4-multimodal
+- phi
+- phi-4-mini
+base_model: microsoft/Phi-4-multimodal-instruct
 ---
 
-# Model Card for Model ID
+# Phi-4-Audio
 
-<!-- Provide a quick summary of what the model is/does. -->
+**Phi-4-Audio** is a highly efficient adaptation of the [Phi-4-multimodal-instruct](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) model, exclusively optimized for audio-text interactions (e.g., Automatic Speech Recognition).
 
+By surgically removing the vision processing components—including the image encoder, vision projection layers, and associated processing logic—we have created a streamlined model that delivers lower memory usage while retaining the original model's powerful audio understanding capabilities.
 
+## Usage & Performance
 
-## Model Details
-
-### Model Description
-
-<!-- Provide a longer summary of what this model is. -->
-
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
+This model is ideal for scenarios where audio processing is the sole modality, such as transcription services, voice assistants, and audio-based QA systems. It is also well-suited for researchers aiming to fine-tune the model specifically for audio tasks without the overhead of unused vision parameters.
 
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
+### Key Improvements
 
-### Model Sources [optional]
+Comparing **Phi-4-Audio** against the original **Phi-4-multimodal-instruct** on a single NVIDIA RTX 5090 GPU:
 
-<!-- Provide the basic links for the model. -->
-
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
+* **Reduced Footprint:** Parameter count reduced by approximately **450 Million**.
+* **Lower VRAM Usage:** Peak inference memory usage reduced by **~10% (0.84 GB saved)**.
+* **Same Audio Performance:** Retains full audio-understanding capabilities while running lighter.
 
 ## Uses
 
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-
-### Direct Use
-
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-
-[More Information Needed]
-
-### Downstream Use [optional]
-
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-
-[More Information Needed]
-
-### Out-of-Scope Use
-
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-
-[More Information Needed]
-
-## Bias, Risks, and Limitations
-
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-
-[More Information Needed]
-
-### Recommendations
-
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-
-## How to Get Started with the Model
-
-Use the code below to get started with the model.
-
-[More Information Needed]
-
-## Training Details
-
-### Training Data
-
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+### Intended Use
 
-[More Information Needed]
+* **Automatic Speech Recognition (ASR):** High-fidelity transcription of spoken audio.
+* **Speech Translation:** Direct speech-to-text translation.
+* **Audio Summarization:** Generating summaries from audio recordings.
+* **Spoken Instruction Tuning:** Fine-tuning on pure audio-text pairs.
 
-### Training Procedure
+### Out of Scope
 
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+-   **Image/Vision Tasks:** This model **cannot** process images. Attempts to pass image inputs will fail or raise errors, as the vision encoders have been stripped.
 
-#### Preprocessing [optional]
+## How to Get Started
 
-[More Information Needed]
+The model is fully compatible with the Hugging Face `transformers` library. You can use it exactly like the original model, but inputting images is not supported.
 
+```python
+import torch
+from torch import nn
+from io import BytesIO
+from urllib.request import urlopen
+from soundfile import read
+from transformers import (
+    AutoModelForCausalLM,
+    AutoProcessor,
+    Phi4MultimodalForCausalLM,
+    Phi4MultimodalModel,
+)
 
-#### Training Hyperparameters
 
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+class StrippedVisionModule(nn.Module):
+    def __init__(self):
+        super().__init__()
 
-#### Speeds, Sizes, Times [optional]
+    def forward(
+        self,
+        **kwargs,
+    ):
+        raise ValueError("Vision is not supported")
 
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
 
-[More Information Needed]
+def strip_vision_inplace(
+    model: Phi4MultimodalForCausalLM | Phi4MultimodalModel,
+) -> Phi4MultimodalForCausalLM | Phi4MultimodalModel:
+    passed_model = model
 
-## Evaluation
+    if isinstance(model, Phi4MultimodalForCausalLM):
+        model = model.model
 
-<!-- This section describes the evaluation protocols and provides the results. -->
+    emb_ext = model.embed_tokens_extend
+    if hasattr(emb_ext, "image_embed"):
+        emb_ext.image_embed = StrippedVisionModule()
+    if hasattr(emb_ext.audio_embed, "down_proj_for_vision_speech"):
+        emb_ext.audio_embed.down_proj_for_vision_speech = StrippedVisionModule()
+    if hasattr(emb_ext.audio_embed, "up_proj_for_vision_speech"):
+        emb_ext.audio_embed.up_proj_for_vision_speech = StrippedVisionModule()
 
-### Testing Data, Factors & Metrics
+    try:
+        torch.cuda.empty_cache()
+    except:
+        pass
 
-#### Testing Data
+    return passed_model
 
-<!-- This should link to a Dataset Card if possible. -->
 
-[More Information Needed]
+model_path = "JacobLinCool/phi-4-audio"
+device = "cuda"
+model = AutoModelForCausalLM.from_pretrained(
+    model_path, device_map=device, dtype=torch.bfloat16
+)
+processor = AutoProcessor.from_pretrained(model_path)
+strip_vision_inplace(model)
 
-#### Factors
 
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+audio_url = "https://huggingface.co/datasets/JacobLinCool/audio-testing/resolve/main/audio/audio-1.mp3"
+audio, samplerate = read(BytesIO(urlopen(audio_url).read()))
 
-[More Information Needed]
+user_prompt = "<|user|>"
+assistant_prompt = "<|assistant|>"
+prompt_suffix = "<|end|>"
+speech_prompt = "Transcribe the audio clip into text."
+prompt = f"{user_prompt}<|audio|>{speech_prompt}{prompt_suffix}{assistant_prompt}"
 
-#### Metrics
+inputs = processor(
+    text=prompt, audio=[audio], sampling_rate=16000, return_tensors="pt"
+).to(device)
 
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+generate_ids = model.generate(**inputs)
+response = processor.batch_decode(
+    generate_ids[:, inputs["input_ids"].shape[1] :], skip_special_tokens=True
+)[0]
 
-[More Information Needed]
+print(f"{response=}")
+```
 
-### Results
-
-[More Information Needed]
-
-#### Summary
-
-
-
-## Model Examination [optional]
-
-<!-- Relevant interpretability work for the model goes here -->
-
-[More Information Needed]
-
-## Environmental Impact
-
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-
-## Technical Specifications [optional]
-
-### Model Architecture and Objective
-
-[More Information Needed]
-
-### Compute Infrastructure
-
-[More Information Needed]
-
-#### Hardware
-
-[More Information Needed]
-
-#### Software
-
-[More Information Needed]
-
-## Citation [optional]
-
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-
-**BibTeX:**
-
-[More Information Needed]
-
-**APA:**
+## Model Details
 
-[More Information Needed]
+- Base Architecture: Phi-4 Multimodal
+- Modifications:
+  - Removed `embed_tokens_extend.image_embed`
+  - Removed `audio_embed.down_proj_for_vision_speech`
+  - Removed `audio_embed.up_proj_for_vision_speech`
 
-## Glossary [optional]
+## Comparisons
 
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+### Parameter Count
 
-[More Information Needed]
+| Model | Total Parameters | Reduction |
+| --- | --- | --- |
+| Phi-4-multimodal-instruct | 4,743,988,032 (4.74B) | - |
+| Phi-4-Audio | 4,289,848,960 (4.29B) | -454M |
 
-## More Information [optional]
+### Benchmark Results (A40)
 
-[More Information Needed]
+Tested on NVIDIA RTX 5090, `torch.bfloat16`.
 
-## Model Card Authors [optional]
+| Metric | Original Model | Phi-4-Audio | Delta |
+| --- | --- | --- | --- |
+| Peak Memory (GB) | 8.88 GB | 8.04 GB | -0.84 GB |
+| Inference Speed (Warm) | ~100.5 tokens/s | ~100.5 tokens/s | Similar |
 
-[More Information Needed]
+## Citation
 
-## Model Card Contact
+If you use this model version, please cite the original Phi-4 Multimodal paper and acknowledge the modifications.
 
-[More Information Needed]
+```bibtex
+@article{abouelenin2025phi,
+  title={Phi-4-mini technical report: Compact yet powerful multimodal language models via mixture-of-loras},
+  author={Abouelenin, Abdelrahman and Ashfaq, Atabak and Atkinson, Adam and Awadalla, Hany and Bach, Nguyen and Bao, Jianmin and Benhaim, Alon and Cai, Martin and Chaudhary, Vishrav and Chen, Congcong and others},
+  journal={arXiv preprint arXiv:2503.01743},
+  year={2025}
+}
+```