https://github.com/arnavballincode/realtime_audiostream_cn

https://github.com/arnavballincode/realtime_audiostream_cn

Last synced: 10 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/arnavballincode/realtime_audiostream_cn
• Owner: ArnavBallinCode
• Created: 2025-08-08T13:11:24.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2025-08-09T18:15:45.000Z (11 months ago)
• Last Synced: 2025-08-09T20:27:28.463Z (11 months ago)
• Language: Python
• Size: 5.86 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Real-Time Voice Streaming over UDP

This project demonstrates real-time audio streaming from one computer (sender) to another (receiver) over a local network using Python, UDP sockets, and the sounddevice library.

## How It Works

### 1. Sender (`voice_sender.py`)

- Captures audio from the microphone in small chunks (frames) using the sounddevice library.

- Each chunk is converted to bytes and sent as a UDP packet to the receiver's IP address and port.

- Uses UDP for low-latency, real-time transmission (no retransmission, no ordering, no connection setup).

### 2. Receiver (`voice_receiver.py`)

- Listens for incoming UDP packets on a specified port (default: 5005) on all network interfaces.

- For each received packet, the audio data is extracted and played immediately using the sounddevice library.

- If no packet is received within a short timeout, silence is played to avoid audio glitches.

## Network Details

- **Addressing:**

  - The sender targets the receiver's IP address (e.g., `10.0.1.217`).

  - The receiver binds to `0.0.0.0`, meaning it listens on all available network interfaces.

- **Port:**

  - Both sender and receiver use the same UDP port (default: 5005).

- **Protocol:**

  - UDP (User Datagram Protocol) is used for fast, connectionless communication.

  - No guarantee of delivery, order, or duplication protection (ideal for real-time audio).

## Audio Details

- **Sample Rate:** 44100 Hz (CD quality)

- **Channels:** Mono (1 channel)

- **Chunk Size:** 1024 samples per packet

- **Format:** 16-bit signed integers (`int16`)

## How to Run

1. **On the receiver (Windows):**

   - Run `python voice_receiver.py`.

2. **On the sender (Mac):**

   - Set the correct `RECEIVER_IP` in `voice_sender.py` to the receiver's IP.

   - Run `python voice_sender.py`.

## What Happens Internally

- The sender continuously captures and sends audio packets to the receiver.

- The receiver's callback function fetches each UDP packet, converts it to audio, and plays it in real time.

- If a packet is lost or delayed, the receiver fills the gap with silence (no blocking or waiting).

- This setup is unicast (one-to-one), connectionless, and optimized for low-latency streaming.

## Requirements

- Python 3.x

- `sounddevice` and `numpy` libraries (install via `pip install sounddevice numpy`)

- Both devices must be on the same local network and have UDP port 5005 open (check firewall settings).

---

This project is a simple example of real-time audio streaming and can be extended for more advanced use cases (e.g., multi-user, encryption, error correction, etc.).