TTMBA: Towards Text To Multiple Sources Binaural Audio Generation

Please use earphones for a better experience

Text to Mono Audio Generation

Input TangoFlux Make-An-Audio 2 AudioLDM TangoFlux-NFS
A horse neighs followed by horse trotting and snorting
A police car with siren blaring approaches and then recedes
Helicopter blades spinning
Whistling and then a female singing

Mono to Binaural Audio Rendering

Position Grounf Truth NFS NIIRF BinauralGrad Pyroomacoustics
A child whining and crying
Azimuth:45 Elevation:0
Left Front
A toy helicopter flying as wind blows into a microphone
Azimuth:100 Elevation:0
Left Rear
A motorboat engine running as water splashes and
a man shouts followed by birds chirping in the background
Azimuth:245 Elevation:0
Right Rear

Text To Multiple Sources Binaural Audio Generation

Text Input TTMBA
Someone snores from the left rear below followed
by the helicopter blades spins from the right rear
Toilet flushing in the left side below very close to me
followed by a bird chirping in the right side far away