Hello,
I am looking for a highly skilled FFmpeg and audio processing expert to solve a specific and challenging task.
Project Goal:
To create a reusable FFmpeg command-line script that replicates the unique, high-quality audio signature of a reference MP3 file. This is NOT a standard conversion task; it requires deep expertise in audio engineering and digital signal processing (DSP).
The Challenge:
I have reference MP3 files (one music, one dialogue) that were converted by a specific service (ClipConverter). These files have a distinct "warm" and "clean" audio character that is superior to a standard FFmpeg conversion. The goal is to reverse-engineer this audio signature and build a filterchain that can consistently reproduce it.
Key Technical Requirements & Deliverables:
The final deliverable will be a standalone FFmpeg command-line script. The project will be divided into two paid milestones:
Milestone 1: Analysis & Proof-of-Concept (PoC)
1. Analysis Report: Analyze the provided reference MP3 files and deliver a short report with key metrics (e.g., Integrated LUFS, LRA, True Peak, RMS, and general spectral characteristics).
2. PoC Script: Provide an initial FFmpeg filterchain that processes a source video and outputs a mono 64kbps MP3 file matching the following objective targets:
- Loudness: -16 LUFS (±1 dB tolerance)
- True Peak: ≤ -1.5 dBTP
- Loudness Range (LRA): ≤ 8 LU
3. The script must be well-documented, explaining the choice of each filter.
Milestone 2: Final Tuning & Optimization
(To be initiated only after successful completion of Milestone 1)
Refine the initial script with advanced filters (e.g., multi-band compressor, dynamic equalizer) to precisely match the spectral "color" and dynamic character of the reference files.
Ensure the script is robust and performs well across different types of source audio (music, speech).
Required Skills:
- Deep expertise in FFmpeg and its audio filtering library (libavfilter).
- Strong background in Digital Signal Processing (DSP) and Audio Engineering.
- Experience with audio analysis and measurement tools.
- Familiarity with loudness standards (EBU R128, ITU-R BS.1770).
- Excellent problem-solving and reverse-engineering skills.
To Apply:
In your proposal, please briefly mention your experience with complex audio filterchains and loudness normalization. Do not send a generic proposal.
I am looking for a true expert for this task and the budget is negotiable for the right candidate.
Thank you.
... Show more