Write a short 250-500 word reflection on the installation/performance process. What parts were most effective? What were some challenges you had to work through? Did audience members react they way you expected? How did you work together as a group? Include some documentation of the piece as well if you have it. Turn in as a website on the class Glitch page.
One of the reasons the Nether is so poignant is its relevance to modern day ideas about online identity. We want to perform a scene of the nether to show the modern equivalent of the interactions in it: aka concealing your identity and performing another one with puppetry and face filters. In the creation of this, and in picking photos to represent our characters, we questioned what ways people’s bodies are formed in the nether. Are they chosen from a library stock images/models? Are they created purely from the imagination, or from people’s memories?
We faced challenges when it came to avatar creation and text to speech. While we initially invisioned a live system where we could control an avatar speaking through text-to-speech, the technology was too difficult/costly to prototype in this time frame (we checked out Reallusion's tools but got quickly confused and also recognized the limitations in the free version). To create a prototype for the class, we ended up creating a canned, hack-y performance by combining two separate tools, one for generating an image of a face talking, and one for converting text to speech. We used talkr app, a mobile app that converts a photo to a talking video, and a tool called acapella box that happened to have realistic children's and adult voices. When we performed this, it wasn't clear to people that we were actually using these separate tools and hacking them together; some people gave feedback that implied they thought some part of the performance was generated live or interactive, which maybe speaks to how "faking" what we wanted to do actually was successful to some extent. One challenge in actually performing this was using these canned videos to simulate a conversation. We had to practice timing pressing play and pause on our two character videos, since we both used separate video files. We also wanted to "act" in this conversation by trying to simulate typing noises on physical keyboards, but this was challenging because it would make error sounds on the video. We think if we actually made a system to type and generate text-to-speech in real time, this part of the performance would have been more natural.
We worked together on choosing some themes from the class discussion. We thought that the interesection of thresholds and identities was an interesting subject matter, and wanted to explore the interface between people's online identities and their IRL actions. We ran with a sketch of two monitors facing each other, representing two actor's digital avatars. Later we decided on using scene 10 from the Nether between Morris and Iris, because it was the most thematic in terms of characters having different layers of identities. We then researched technological solutions (see above), and split up the work by each generating our own video that we would control in the performance. Olivia did Morris' avatar, and Jackie did Iris' avatar.