Soundraft - Audio Sampler

Soundraft is a browser-based audio sampler and non-linear arranger that enables users to capture and manipulate audio from YouTube videos using the Web Audio API. The extension operates entirely within the browser and does not perform any downloading or media extraction outside the context of user-initiated streaming. Key Capabilities Injection of visible or background YouTube iframes into the active tab to isolate specific video sources for processing Use of the captureStream() API to access the audio output of these video elements in real time Creation of audio buffers in memory for slicing, looping, and timeline-based arrangement Marker placement and tempo tools to support rhythmic alignment and structured playback A multi-track interface that allows segment-level arrangement of buffered audio clips All audio is processed locally and temporarily. No data is downloaded, recorded, or stored permanently. When a session is shared, only metadata (such as video IDs, segment start/end times, and arrangement structure) is transmitted. Playback sessions always re-stream content directly from YouTube. Soundraft is designed for users who wish to experiment with time-based audio editing techniques using publicly available streaming media in a compliant, memory-safe environment. MVP Disclosure This release is a Minimal Viable Product (MVP) intended to validate that the current implementation does not violate any terms of service, particularly those of YouTube or the Chrome Web Store. The extension is still in active development, and not all intended features are implemented at this stage. Current implementation includes: Capturing audio from user-initiated YouTube video playback via the captureStream() API Injecting YouTube iframes into the page for isolated audio capture and manipulation (including teardown logic after buffering to ensure performance) Processing all audio in memory, without downloading, storing, or transmitting any media content Visual interface for arranging and playing back buffered segments Planned Functionalities We welcome review and feedback on the following planned features, which will only be implemented if they can be developed in accordance with relevant platform terms: 1. Uploading Final Results via YouTube API Users may eventually be able to export arranged compositions and upload them as new videos using the official YouTube Data API. This feature would: Require user-authenticated access (OAuth) Only use buffered audio segments from publicly available content Include proper attribution and metadata 2. Sharing of Metadata Projects can be exported and shared through metadata only. A typical project file might look like: json Copier Modifier { "projectName": "demo_mix", "tempo": 120, "markers": [0.0, 1.0, 2.0], "clips": [ { "videoId": "dQw4w9WgXcQ", "start": 12.3, "end": 19.5, "track": 1 }, { "videoId": "kJQP7kiw5Fk", "start": 3.2, "end": 6.8, "track": 2 } ] } Playback using shared metadata still requires full access to YouTube and will always re-stream content from the source. 3. Keyboard and MIDI Mapping (coming soon) Upcoming releases will introduce: Keyboard mapping: Assign clip triggers and transport controls to physical keyboard keys for faster workflows MIDI mapping: Support for mapping hardware MIDI devices to trigger clips, control playback, adjust volume, and interact with other interface elements These features will enhance live performance and hardware integration capabilities within the browser environment. Responsible Usage Soundraft does not promote or support the use of ad blockers when accessing YouTube. Ads are an integral part of YouTube’s content ecosystem and revenue model. To ensure an optimal sampling experience, users are encouraged to use this extension in combination with YouTube Premium, which offers an ad-free experience without violating platform policies. Sampling workflows are particularly sensitive to interruptions, and YouTube Premium ensures that buffered segments are clean and consistent without modifying YouTube’s behavior.