Type any script. Hear it back in that Bronx-drill sleepy-confident speaking register — relaxed mid-alto, half-bored on the surface, the dry uptick at the end of every sentence that signals the punchline is, in fact, the entire setup. Internet-native cadence. Studio-quality MP3 in under a minute. No software to install. Built on HyperVoice, our proprietary neural TTS engine.
Isis Gaston speaks in a sleepy-confident Bronx mid-alto that grew up Fordham-Road-internet-native, learned its public register on TikTok before the radio caught up, and never sanded out the half-bored uptick that defines her cadence. The voice does not perform. The half-bored register is the entire performance — the listener is supposed to understand that nothing being said is interesting enough to push for, and that fact is, somehow, the punchline.
TaskAGI's Ice Spice AI voice generator runs on HyperVoice, our proprietary text-to-speech engine. The model captures that Bronx-drill speaking register specifically — the Fordham-Road baseline, the half-bored uptick on the closing word of each phrase, the internet-native cadence that defines the press-tour mode, and the Gen-Z chronic-online inflection she carries on brand work.
Four presets target modes. Press is the default red-carpet and on-camera register, sleepy-confident with the uptick intact. Conversational warms slightly for podcast-couch reads. Brand tightens for spokesperson and product-launch voiceover. Sardonic brings the internet-native smirk fully forward.
Creators reach for this voice when a script needs Gen-Z internet-native warmth with Bronx specificity underneath. Drill-history documentary cold-opens. Internet-culture YouTube essays. Gen-Z brand voiceover that can't read as corporate. Bronx-music-scene scripts. The voice does work that a generic young-female-rap preset cannot do because it carries a specific learned restraint — the restraint of a person who learned to be famous on TikTok and never gave it up.