Make $25 - $35 Per Hour Doing Simple Jobs From Home.
What is your current level of experience in freelance work?
AI/ML Developer Needed to Build AI Music Sample Generator (Audio + Keyword Input)
We’re developing an AI-powered music platform that generates royalty-free production samples using mood, genre, and style inputs. We’re hiring an experienced AI/ML developer to help build the core audio generation engine for our MVP. About the Project: The platform will let users generate short, high-quality samples (4-bar, 8-bar loops, and one-shots) by selecting up to 3 descriptive keywords such as "dark," "gritty," "ambient," "808-heavy," etc. The goal is to enable fast, intuitive, text-to-audio generation based on genre, emotion, and texture. We’ll be working with a curated, royalty-free sample dataset organized by genre, key, instrument, BPM, and descriptive tags — starting with Hip-hop, Trap, and Drill. Responsibilities: - Fine-tune or adapt a state-of-the-art audio generation model (e.g., Stable Audio) - Implement keyword-driven generation logic using prompt embeddings or conditioning - Generate short-form, loopable, and production-ready .wav audio - Ensure clean .wav export in 44.1kHz for DAW compatibility - Optimize model for inference speed and generation quality - Deliver an inference API or script (e.g., Flask, CLI, or cloud-deployable endpoint) - Integrate a tagging/embedding model like CLAP for searchable categorization Preferred Skills: - Strong experience with Python and PyTorch - Familiarity with audio generation models: MusicGen, Stable Audio, RAVE, Jukebox, etc. - Experience fine-tuning generative models on small/mid-size datasets - Understanding of audio embedding tools (CLAP, Wav2Vec) - Background in DSP, music production, or generative music tools is a big plus Deliverables: - Fully functional fine-tuned model trained on our dataset - Output: .wav samples (loops + one-shots), triggered via up to 3 user-defined keywords - JSON or metadata return for tag and prompt tracking - Documented usage instructions for API or CLI inference Timeline: - Goal: MVP within 4–6 weeks - Potential for long-term work (full deployment, ongoing development, or CTO role) Budget: - We’re open to fixed milestone based compensation. Please include: - Your availability and proposed development timeline - Relevant past projects (especially audio/music AI) - Your preferred model and why you'd use it for this application This is an exciting opportunity to work on the cutting edge of music and generative AI. If you’ve built models that generate, tag, or manipulate music/audio — we want to hear from you.