↓ 1 callersMethoddecodeDecode acoustic latents to audio waveform. Input: (batch, vae_dim, frames) or (batch, frames, vae_dim) Output: (batch, 1, samples)
cake-core/src/models/vibevoice/vae_decoder.rs:399
↓ 1 callersMethoddepthwise_conv1d_bias(&self, i: &Tensor, w: &Tensor, b: &Tensor, ks: usize, _: usize)
cake-core/src/backends/rocm/mod.rs:404
↓ 1 callersMethoddepthwise_conv1d_bias(&self, padded_input: &Tensor, weight: &Tensor, bias: &Tensor, _kernel_size: usize, _channels: usize)
cake-core/src/backends/metal/mod.rs:884
↓ 1 callersMethoddepthwise_conv1d_bias(
&self,
input: &Tensor,
weight: &Tensor,
bias: &Tensor,
ks: usize,
cake-core/src/backends/vulkan/mod.rs:1435