# aeria is human-centric music discovery, augmented by visualized AI.

“they took my mixtape off soundcloud kuz it was ‘doin too good’ n politics...i miss music minus the politics”. ![image.png](https://cdn.steemitimages.com/DQmXRePBacheM5NLcsqaDZ5Cfg93d1V6gVuMbTzBqfd5XRY/image.png) Revered underground artist Smino’s quote echoes a sentiment the music community has long felt deeply: despite the explosion in music availability and information it provides, the streaming era has corrupted the cultural experience of music discovery and community. artists sign their creative and financial lives away to labels in exchange for creating fake hype cycles around their work through insider deals and thinly-veiled pay-for-play marketing schemes. once processed and sterilized for mass market consumption, that work then reaches the community via elitist playlisting and exploitative algorithms. the organic process of connecting to and through music, especially online, feels like a cultural artifact. we need a way for data to make the cultural exchange of music better, not worse, and for all creators and their communities, not simply a few major platforms and rights holders. # visualized music intelligence
aeria is a free and open standard for exploring music and music communities through immersive music data visualization. it puts powerful AI at the service of every creator, enabling them to independently discover the support and community they need to sustain their work. aeria visualizes content and social networks to display them immersive, fluid visual music feature spaces instead of rigid lists or grids. In these feature spaces, visual cues symbolize human-meaningful insights: for instance, symbolizing * artists as balls, and ties between artists, such as social media follows or collabs, as lines connecting the balls * relatedness with closeness, so that communities of related artists form clusters of balls appearing as distinct from other clusters * *influence*, not popularity, with ball size we can adapt this basic symbolism to different skins, or visualization styles. for instance, visualizing artists as stars in a night sky, ties as constellations, and communities as galaxies. ![image.png](https://cdn.steemitimages.com/DQmT1TXToycgC7rhZtoxSZXWNNWdPnMcEqVJrLNssmkYQh6/image.png) [wikiverse: a galactic reimagining of wikipedia](http://wikiverse.io/), imaged above, is a great reference here, with aeria visualizing the space of songs and artists rather than wiki articles. # machine music knowledge
generally, the subjective experience of music or any media depends on two main factors: * content, or the perceived qualities of the sound, such as its tempo, key, mode, rhythm, timbre, etc * context, eg the social, cultural, historical, political, or even psychological conditions under which one listens to music aeria takes both into account. it understands content via audio analysis based on digital signals processing. it learns about context largely via social network analysis done on social media data. in visually capturing its combined, interwoven knowledge of musical content and context, aeria closes the chasm between artists and their potential community, from both a sonic and social perspective. # go-to-market strategy ## integrations
aeria is an open-source standard and will integrate with any platform from which we can acquire data valuable to our participants. at our [github](https://github.com/afuturemodern/aeria), you can view code for pulling data from soundcloud’s API to construct an visualize graph its users, which we have done successfully on our local machines. we also feel twitter and youtube would make powerful integrations in particular, as these are all free services focusing on free to view, user-generated content from generally anonymous profiles, making them great for gathering insight about the tastes and interests of different communities without violating privacy. ## partnerships to further bootstrap our database and our participation, aeria seeks direct partnership with more community-led or decentralized music services, to provide functionality directly through their platform. so far, we have agreed in principle to partnerships with [Audius](https://audius.co/afuturemodern) and [Resonate](https://resonate.is/profile/3048/), and have discussions ongoing with [current.us](https://current.us/), [yungcloud](https://yung.cloud/), [PopChest](https://thepopnetwork.org/), [currents.fm](https://a.currents.fm/) and more. # use cases
several potential use cases for aeria and future modern’s initiatives in music excite us. ## multi-dimensional creator dashboard
some services out there charge artists to access the bare minimum of their own data. none freely and fully unpack music data to unlock all its potential for artists. aeria will go beyond tracking followers and week-over-week growth, showing * artists their specific individual likely fans based on their taste, not their demographic * vocalists what beats / producers fit the keys they tend to sing in * composers which stem will add the finishing touch to their piece based on its key and tempo and much, much, more, all through immersive, dynamic visualization of soundspace, enabling artists to self-guide their creative path augmented by AI, rather than the AI deciding their steps for them. ## immersive media navigation guide
rather than scroll through songs in a detached, disconnected feed, explore a virtual music space that expresses relationships in music through the placing of content in this space. content displayed closer together will tend to have similar moods. this enables musical category to arise from the bottom up based on our every day experience of sound and culture, rather than top down by industry “experts” based on outdated notions of genre. in fact, participants can describe the mood of a cluster with their own tags, eg “enigmatic trap” or “triumphant bass music”, giving rise to crowdsourced, ai-coordinated genre classification. put the human back in the algorithm, and make discovering music fun again. ## creator tools: smart audiovisual workstation
ai can also augment in the creative process itself, without replacing human ingenuity and curiosity. we see this looking like * light shows synchronizing and changing the color of the lights based on extracted midi information from isolated instrument tracks * smart marketplaces for both human and ai-made synths for customizing FPGA-based synthesizers and more. ## music’s (de)central nervous system
ultimately, we see aeria as the brain of the future decentralized music ecosystem. we leverage smart services to unbundle the strategic side of music, empowering every artist to inform their own strategy without a record label. this way, artists retain more of their profits by signing better deals or not needing them in the first place, so they can focus on their art instead of how to sell it. |record label function|unbundled service| |-|-| |

promotion: playlisting, radio plugging, global markets

partnership: sync, brand

publicity: newspapers, interviews, blogs

|social network visualization service that

Visualizes the social, cultural, and emotional structure of the music ecosystem, through a hierarchical clustering of content and communities, ranging from cultures at the highest level, to subcultures, scenes, niches, down to cliques and collectives

effectively communicates what roles people play within the ecosystem by represents different kinds of profiles with differently shaped icons, eg megaphone for promoter, parchment for newspaper, etc

conveys each person’s relevance or importance to their respective community by sizing these icons proportionally to their associated profiles’ influence metric

coaches artists in how to network with these influencers by highlighting shortest connected path between any two profiles