load tts tokenizer and vocos 1. try load form local 2. try load from huggingface

Merge `input_ids` and `lm_spk_emb_last_hidden_states` to `inputs_embeds`. Args: input_ids (torch.Tensor): Input token IDs.

Prefill a chunk of audio ids to the model. Used in sliding-window long audio generation. Specifically, prefill many audio ids (typically from

:param image: shape [3, H, W] :param patch_size: :return: [3, patch_size, HW/patch_size]

Create mask for subsequent steps (size, size) with chunk size, this is for streaming encoder Args: size (int): size of ma

Converts `image` to a PIL Image. Optionally rescales it and puts the channel dimension back as the last axis if needed. Args

Fills the input Tensor with values drawn from a truncated normal distribution. The values are effectively drawn from the normal distribution :

audio: Tensor([num_channels, num_samples])