Upload 4 files

.gitattributes

@@ -1,35 +1,35 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,15 +1,42 @@
----
-title: GTROX
-emoji: 💬
-colorFrom: yellow
-colorTo: purple
-sdk: gradio
-sdk_version: 6.5.1
-app_file: app.py
-pinned: false
-hf_oauth: true
-hf_oauth_scopes:
-- inference-api
----
-
-An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).
+---
+title: GTROX
+emoji: 💬
+colorFrom: yellow
+colorTo: purple
+sdk: gradio
+sdk_version: 6.5.1
+app_file: app.py
+pinned: true
+hf_oauth: true
+hf_oauth_scopes:
+- inference-api
+---
+
+# 🎙️ GTROX: Speech Insight Coach
+
+**GTROX** is an AI-powered speech analysis tool designed to help you improve your fluency, pacing, and clarity. By leveraging advanced speech-to-text models and compact, high-performance Language Models (SLMs), it provides actionable, real-time coaching feedback based on your audio input.
+
+## ✨ Key Features
+
+* **Fluency Tracking:** Instantly calculate your Words Per Minute (WPM) to master your speaking pace.
+* **Silence Analysis:** Identify how much of your speech consists of pauses and filler time.
+* **AI Coaching:** Get personalized, constructive tips generated by a specialized Language Model to improve your delivery.
+* **Robust Pipeline:** Engineered for high performance, utilizing efficient audio processing without heavy external dependencies.
+
+## 🛠️ How It Works
+
+
+
+1. **Input:** The system processes your uploaded `.wav` file or live microphone recording.
+2. **Analysis:** It calculates technical metrics like WPM and silence ratios using numerical analysis.
+3. **Transcription:** Uses `distil-whisper` for highly accurate, low-latency speech-to-text.
+4. **Feedback:** The `SmolLM2` language model analyzes your performance and suggests improvements.
+
+## 🚀 Deployment
+
+This app is built with **Gradio** and hosted on **Hugging Face Spaces**. It is optimized for low-latency inference, making it suitable for quick fluency checks on the go.
+
+## 📦 Requirements
+
+To run this application locally, ensure you have Python 3.10+ and the necessary dependencies installed:
+

app.py

@@ -1,69 +1,82 @@
-import gradio as gr
-from huggingface_hub import InferenceClient
-
-
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
-    """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-    """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
-
-    messages = [{"role": "system", "content": system_message}]
-
-    messages.extend(history)
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-chatbot = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
-
-with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
-
-
-if __name__ == "__main__":
-    demo.launch()
+import gradio as gr
+import librosa
+import numpy as np
+import torch
+import soundfile as sf
+from transformers import pipeline
+import os
+
+# 1. Model Initialization
+# We set to 'cpu' as default for broader compatibility, but it will detect 'cuda' if available.
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+# Transcriber: Fast, efficient speech-to-text
+transcriber = pipeline("automatic-speech-recognition", 
+                       model="distil-whisper/distil-small.en", 
+                       device=device,
+                       return_timestamps=True
+                       )
+
+# Coach: Small Language Model for feedback
+coach = pipeline("text-generation", 
+                 model="HuggingFaceTB/SmolLM2-1.7B-Instruct", 
+                 device=device)
+
+def analyze_speech(audio_path):
+    if audio_path is None:
+        return "Please upload an audio file."
+
+    # 1. Load Data
+    try:
+        data, sr = sf.read(audio_path)
+        if len(data.shape) > 1: data = data.mean(axis=1)
+        if sr != 16000:
+            data = librosa.resample(data, orig_sr=sr, target_sr=16000)
+    except Exception as e:
+        return f"Error loading audio: {str(e)}"
+
+    # 2. Transcription (Using optimized chunking)
+    # This prevents the "long-form" 30s crash
+    try:
+        # Pass the array and sample rate directly
+        result = transcriber({"raw": data, "sampling_rate": 16000})
+        text = result["text"]
+    except Exception as e:
+        return f"Transcription error: {str(e)}"
+
+    # 3. Metrics
+    duration = len(data) / 16000
+    wpm = (len(text.split()) / (duration / 60)) if duration > 0 else 0
+    silence_ratio = np.sum(np.abs(data) < 0.01) / len(data)
+
+    # 4. Feedback
+    prompt = (
+        f"You are a professional speech coach. The user spoke at {wpm:.1f} WPM with {silence_ratio:.1%} silence. "
+        f"Transcript: '{text}'. "
+        "Provide exactly 2 short, constructive tips to improve their delivery, focusing on pacing and clarity."
+    )
+    try:
+        # Simplified generation call
+        feedback = coach(prompt, max_new_tokens=100)[0]['generated_text']
+    except Exception as e:
+        feedback = "Could not generate feedback."
+
+    return f"📝 Transcription: {text}\n\n📊 Metrics: {wpm:.1f} WPM, Silence: {silence_ratio:.1%}\n\n💡 Coach Feedback:\n{feedback}"
+
+
+# 2. UI Layout
+with gr.Blocks() as demo:
+    gr.Markdown("# 🎙️ ImproveTalk: Speech Insight Coach")
+    gr.Markdown("Upload your `.wav` file to get instant feedback on your fluency.")
+    
+    with gr.Row():
+        # audio_input = gr.Audio(type="filepath", label="Upload WAV Audio")
+        audio_input = gr.Audio(type="filepath", label="Record or Upload Audio",sources=["microphone", "upload"])
+        analyze_btn = gr.Button("Analyze My Speech", variant="primary")
+    
+    results = gr.Textbox(label="Analysis Results", lines=10)
+    
+    analyze_btn.click(analyze_speech, inputs=audio_input, outputs=results)
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1,16 @@
+# UI Framework
+gradio>=5.0.0
+
+# Core AI/ML Libraries
+torch>=2.0.0
+transformers>=4.40.0
+accelerate>=0.30.0
+
+# Audio Processing
+librosa>=0.10.0
+soundfile>=0.12.0
+numpy>=1.26.0
+
+# Inference Optimization (Crucial for Small Models)
+bitsandbytes>=0.43.0
+