Audio Gain Control (AGC - WebRTC)
Overview
This project provides a simple tool for processing audio files with features like:
- Automatic Gain Control (AGC): Dynamically adjusts the volume of the audio to ensure consistent levels using WebRTC’s GainController2.
- Custom Resampling: Converts audio to a desired sample rate (e.g., 48000 Hz) - using linear interpolation for resampling.
- Flexible Input/Output: Supports reading and writing raw audio files, with optional integration of FFmpeg for advanced audio handling.
How It Works
- Audio Input:
- Reads raw PCM audio from a file or uses FFmpeg for format decoding.
- Processing:
- Applies AGC using WebRTC’s GainController2.
- Optionally resamples the audio to a specified sample rate.
- Audio Output:
- Writes processed audio to a raw PCM file or uses FFmpeg for encoding.
Build Instructions
-
Ensure you have a C++ compiler that supports C++17 or later and cmake installed.
-
Clone the repository:
git clone <repository_url> cd <repository_folder> -
Initialize the thirdparty agc libs and make sure to have meson and ninja installed.
cd thirdparty sh install.sh -
Check the agc/libs folder consist of .so or .a files. If they are present inside x86_64-linux-gnu folder, get them out and place them in the agc/libs folder. Once done,you can go back to the root folder and continue with the build process.
-
Compile the code:
cmake -S . -B build cmake --build build
Running the Program
To process an audio file, use the following command:
./bin/audioProcessor <input_file> <output_file>
<input_file>: Path to the raw PCM input file (with a WAV header).<output_file>: Path to the processed raw PCM output file.
Dependencies
- C++17: For modern C++ features.
- CMake: For building the project.
- Meson and Ninja: For building the thirdparty agc libs.
- WebRTC: For GainController2 and audio buffer management.
- FFmpeg (optional): For reading and writing non-raw audio formats.
Contributing
Feel free to open issues or submit pull requests for enhancements or bug fixes.