r""" This is the configuration class to store the configuration of a [`FunAudioChatEncoder`]. It is used to instantiate a Qwen2-Audio audio en

Multi-headed attention from 'Attention Is All You Need' paper

Transformer encoder consisting of *config.encoder_layers* self attention layers. Each layer is a [`FunAudioChatDecoderLayer`]. Args:

Custom Streamer for streaming text_ids and audio_ids The model will call put() twice: 1. streamer.put(next_tokens) - Text tokens

r""" This is the configuration class to store the configuration of a [`FunAudioChatForConditionalGeneration`]. It is used to instantiate an Qw

FunAudioChat multimodal plugin for audio, image and video processing.

r""" Constructs a FunAudioChat processor which wraps a FunAudioChat feature extractor and a FunAudioChat tokenizer into a single processor. [