Chimera Vocal Synth

269 points · 44 comments on HN · read original →

Points and comments are a snapshot, not live.

Chimera simulates speech by modeling air through a throat, mouth, and nose.

Chimera is a web-based vocal synthesizer that produces sound by simulating airflow through a modeled vocal tract. Users can drag the tongue, lips, or jaw, and play keyboard keys to change pitch. The interface includes controls for volume, breath pressure, pressed quality, vibrato, and pitch (default 110 Hz A2). The 3D view shows sound pressure and airflow. The tool runs in a browser and is open source on GitHub.

What commenters are saying

Commenters compared Chimera to Pink Trombone and the 1939 Bell Labs Voder. Trans users saw potential for voice affirmation training. One user successfully produced Khoomei overtone singing. Others created multi-part harmonies (e.g., barbershop quartet) and noted a bug causing screeching with extreme settings. Some found the 3D view creepy. The name may reference Teto (Vocaloid).