Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

https://github.com/codecat1111/gemini-interview-assistant

A web-based tool that helps users prepare and practice for technical interviews using AI-powered feedback.
https://github.com/codecat1111/gemini-interview-assistant

assistant gemini interview whisper

Last synced: 28 days ago
JSON representation

A web-based tool that helps users prepare and practice for technical interviews using AI-powered feedback.

Awesome Lists containing this project

README 

          
# 🚀 Gemini Powered Mock Interview Assistant



A web-based tool that helps users prepare and practice for technical interviews using AI-powered feedback. 🎯💡



![Gemini-Interview-Assistant](assets/images/Gemini_Interview_Assistant.png)

---



## 🌟 Features



✅ **Real-time audio recording & transcription** 🎤📝  

✅ **AI-powered analysis of interview responses** 🤖📊  

✅ **Two modes of operation:**



- 🎯 **AI Analysis Only**: Get immediate feedback 🔥

- 🔄 **Compare Mode**: Practice your answer first, then compare with AI analysis 

- ✅ **Session management with history tracking**  

- ✅ **Support for role-specific & job description context** 👨‍💻📄



---



## 🔧 Prerequisites



🛠 **System Requirements:**



-  Python **3.8+**

-  **PyTorch & TorchAudio**

-  **Transformers library**

-  **Google Cloud API key for Gemini AI**

-  **Virtual audio cable (VB-CABLE)**

-  **Internet connection for AI services**

-  **Windows OS (for VB-CABLE setup)**



---



## 🏗️ Setup



### 1️⃣ Clone the Repository



```bash

 git clone https://github.com/codecat1111/Gemini-Interview-Assistant.git

 cd Interview-Assistance

```



### 2️⃣ Install Dependencies



```bash

pip install -r requirements.txt

```



### 3️⃣ Configure Virtual Audio Cable 🎧



- Install a **virtual audio cable** software

- Set **audio output** to "CABLE Input"



### 4️⃣ Set Up API Key 🔑



```bash

# Create .env file and add your Gemini API key

echo "GOOGLE_API_KEY=your_gemini_api_key_here" > .env

```



---



## 🚀 Usage



1️⃣ **Start the Gemini Analyzer Server** (in a new terminal):



```bash

python gemini_analyzer.py

```



2️⃣ **Start the Flask Server** 🌐:



```bash

python main.py

```



3️⃣ **Open in Browser**: `http://localhost:5000` 🌍



4️⃣ **Enter job role & description** ✍️



5️⃣ **Choose Your Mode** 🎙️:



- **AI Analysis Only**: Click "Start Recording" 📢

- **Compare Mode**: Record your answer, then click "Show AI Analysis" 🧐



---



## 🔎 Features in Detail



### 🎙️ **Recording Controls**



- **Start/Stop Recording** button ⏺️

- **New Recording** button for multiple attempts 🔄

- **Reset Session** button to clear history 🗑️



### 📊 **Analysis Options**



- **Real-time transcription** 

- **AI-powered feedback** 

- **Compare Mode for self-assessment** 

- **Session history tracking** 



### 📜 **History Feature**



- 🆕 "Start Recording (No History)" - Begins a new recording without considering previous responses

- 🔄 "New Recording (With History)" - Saves the current analysis to history before starting a new recording

- 📝 **History is automatically included in AI analysis for context**

- 🗑️ **Reset Session button clears all history**



### 🎨 **Theme Support**



- 🌞 **Light Mode** / 🌙 **Dark Mode** toggle

- 🎨 **Theme preference is saved automatically**

-  **Accessible color schemes for better readability**



## Project Demo



[📽️ Watch the Demo on YouTube](https://www.youtube.com/watch?v=gU2my0yqVQY)



This video demonstrates:



- Setting up role and job description

- Recording and analyzing responses

- Using AI Analysis Only mode

- Using Compare mode

- Managing interview history

- Theme switching



---



## 🎤 Model Setup



###  **Loading the Whisper Model**



The application uses the **Whisper-Tiny** model for speech transcription. On first run, the model will be **automatically downloaded** from Hugging Face.



```python

# Initialize Whisper model

processor = AutoProcessor.from_pretrained("openai/whisper-tiny")

model = AutoModelForSpeechSeq2Seq.from_pretrained("openai/whisper-tiny")

```



---



## 🎧 Virtual Audio Cable Setup



### 1️⃣ **Setting Up VB-Audio Virtual Cable**



This project uses **VB-Audio Virtual Cable** to capture system audio and process live transcriptions.



#### 🔽 **Installation**



- Download [VB-Audio Virtual Cable](https://vb-audio.com/Cable/)

- Run `VBCABLE_Setup_x64.exe` (for 64-bit systems)

- Restart your computer 🔄



#### 🎚 **Configuration**



1. Open **Sound Settings** (Win + I → "Sound")

2. **Set "CABLE Input (VB-Audio Virtual Cable)" as default output**

3. **Select "CABLE Output (VB-Audio Virtual Cable)" as microphone input**

4. (Optional) Use **VoiceMeeter Banana** for advanced routing 🎛️



---



## 🎤 System Audio Capture Implementation



### 🎧 **Virtual Cable Device Detection**



```python

def find_vb_audio_device():

    """Finds VB-Audio Virtual Cable device ID"""

    devices = sd.query_devices()

    for i, device in enumerate(devices):

        if "CABLE Output" in device["name"]:

            return i

    return None

```



### 🎙 **Audio Recording System**



```python

class SystemAudioRecorder:

    def __init__(self):

        self.sample_rate = 16000

        self.recording = False

        self.audio_data = []



    def start_recording(self):

        device_id = find_vb_audio_device()

        if device_id is None:

            raise RuntimeError("VB-Audio Virtual Cable not found!")



        self.recording = True

        self.audio_data = []



        def audio_callback(indata, frames, time, status):

            if self.recording:

                self.audio_data.append(indata.copy())



        self.stream = sd.InputStream(

            device=device_id,

            channels=1,

            samplerate=self.sample_rate,

            callback=audio_callback

        )

        self.stream.start()



    def stop_recording(self):

        self.recording = False

        self.stream.stop()

        self.stream.close()

        audio_data = np.concatenate(self.audio_data, axis=0)

        filename = "recorded_audio.wav"

        sf.write(filename, audio_data, self.sample_rate)

        return filename

```



---



## 🛠 Troubleshooting Virtual Audio Cable



⚠ **No Audio Detected?**



- ✅ Verify **"CABLE Input"** is set as **default playback device**

- ✅ Check **"CABLE Output"** is selected as **recording input**

- ✅ Ensure **system volume is not muted**



⚠ **Device Not Found?**



- 🔄 Restart the **VB-Audio service**

- 🔄 Reinstall the **drivers**

- 🔍 Check **Device Manager** for warning icons



⚠ **Poor Audio Quality?**



- 🎚 Ensure **sample rate settings match (16000 Hz)**

- 🔍 Check **CPU usage**

- 🚀 Close **unnecessary audio applications**



## Virtual Audio Cable Setup Demo



[📽️ Watch the Audio Cable Setup on YouTube](https://www.youtube.com/watch?v=7wr8EVFtABM)



This video covers:



- Installing VB-Cable

- Configuring Windows audio settings

- Testing the virtual audio setup

- Troubleshooting common issues



---



## 💡 Contributing



1️⃣ **Fork the Repository** 🍴



2️⃣ **Create a Feature Branch** (`git checkout -b feature/AmazingFeature`)



3️⃣ **Commit Your Changes** (`git commit -m 'Add AmazingFeature'`)



4️⃣ **Push to Branch** (`git push origin feature/AmazingFeature`)



5️⃣ **Open a Pull Request** 



---



## ⭐ Found It Helpful? [Star It!](https://github.com/codecat1111/Gemini-Interview-Assistant/stargazers) ⭐



---



## 📜 License



This project is licensed under the **MIT License** - see the [LICENSE](LICENSE) file for details.