metadata
license: apache-2.0
datasets:
- chenjoya/Live-CC-5M
- chenjoya/Live-WhisperX-526K
- lmms-lab/LLaVA-Video-178K
language:
- en
base_model:
- Qwen/Qwen2-VL-7B
tags:
- qwen_vl
- video
- real-time
- multimodal
- LLM
LiveCC-7B-Instruct
Introduction
We introduce LiveCC, the first video LLM capable of real-time commentary, trained with a novel video-ASR streaming method, SOTA on both streaming and offline benchmarks.
- Project Page: https://showlab.github.io/livecc
This is the SFT model. The base model is at LiveCC-7B-Base.
Training with Streaming Frame-Words Paradigm
Quickstart
Like qwen-vl-utils, we offer a toolkit to help you handle various types of visual input more conveniently, especially on video streaming inputs. You can install it using the following command:
pip install qwen-vl-utils livecc-utils
Here we show a code snippet to show you how to do real-time video commentary with transformers and the above utils:
import functools, torch, os, tqdm
from liger_kernel.transformers import apply_liger_kernel_to_qwen2_vl
apply_liger_kernel_to_qwen2_vl() # important. our model is trained with this. keep consistency
from transformers import Qwen2VLForConditionalGeneration, AutoProcessor, LogitsProcessor, logging
from livecc_utils import prepare_multiturn_multimodal_inputs_for_generation, get_smart_resized_clip, get_smart_resized_video_reader
from qwen_vl_utils import process_vision_info
class LiveCCDemoInfer:
VIDEO_PLAY_END = object()
VIDEO_PLAY_CONTINUE = object()
fps = 2
initial_fps_frames = 6
streaming_fps_frames = 2
initial_time_interval = initial_fps_frames / fps
streaming_time_interval = streaming_fps_frames / fps
frame_time_interval = 1 / fps
def __init__(self, model_path: str = None, device_id: int = 0):
self.model = Qwen2VLForConditionalGeneration.from_pretrained(
model_path, torch_dtype="auto",
device_map=f'cuda:{device_id}',
attn_implementation='flash_attention_2'
)
self.processor = AutoProcessor.from_pretrained(model_path, use_fast=False)
self.streaming_eos_token_id = self.processor.tokenizer(' ...').input_ids[-1]
self.model.prepare_inputs_for_generation = functools.partial(prepare_multiturn_multimodal_inputs_for_generation, self.model)
message = {
"role": "user",
"content": [
{"type": "text", "text": 'livecc'},
]
}
texts = self.processor.apply_chat_template([message], tokenize=False)
self.system_prompt_offset = texts.index('<|im_start|>user')
self._cached_video_readers_with_hw = {}
@torch.inference_mode()
def live_cc(
self,
query: str,
state: dict,
max_pixels: int = 384 * 28 * 28,
default_query: str = 'Please describe the video.',
do_sample: bool = False,
repetition_penalty: float = 1.05,
streaming_eos_base_threshold: float = None,
streaming_eos_threshold_step: float = None,
**kwargs,
):
"""
state: dict, (maybe) with keys:
video_path: str, video path
video_timestamp: float, current video timestamp
last_timestamp: float, last processed video timestamp
last_video_pts_index: int, last processed video frame index
video_pts: np.ndarray, video pts
last_history: list, last processed history
"""
# 1. preparation: video_reader, and last processing info
video_timestamp, last_timestamp = state.get('video_timestamp', 0), state.get('last_timestamp', -1 / self.fps)
video_path = state['video_path']
if video_path not in self._cached_video_readers_with_hw:
self._cached_video_readers_with_hw[video_path] = get_smart_resized_video_reader(video_path, max_pixels)
video_reader = self._cached_video_readers_with_hw[video_path][0]
video_reader.get_frame_timestamp(0)
state['video_pts'] = torch.from_numpy(video_reader._frame_pts[:, 1])
state['last_video_pts_index'] = -1
video_pts = state['video_pts']
if last_timestamp + self.frame_time_interval > video_pts[-1]:
state['video_end'] = True
return
video_reader, resized_height, resized_width = self._cached_video_readers_with_hw[video_path]
last_video_pts_index = state['last_video_pts_index']
# 2. which frames will be processed
initialized = last_timestamp >= 0
if not initialized:
video_timestamp = max(video_timestamp, self.initial_time_interval)
if video_timestamp <= last_timestamp + self.frame_time_interval:
return
timestamps = torch.arange(last_timestamp + self.frame_time_interval, video_timestamp, self.frame_time_interval) # add compensation
# 3. fetch frames in required timestamps
clip, clip_timestamps, clip_idxs = get_smart_resized_clip(video_reader, resized_height, resized_width, timestamps, video_pts, video_pts_index_from=last_video_pts_index+1)
state['last_video_pts_index'] = clip_idxs[-1]
state['last_timestamp'] = clip_timestamps[-1]
# 4. organize to interleave frames
interleave_clips, interleave_timestamps = [], []
if not initialized:
interleave_clips.append(clip[:self.initial_fps_frames])
interleave_timestamps.append(clip_timestamps[:self.initial_fps_frames])
clip = clip[self.initial_fps_frames:]
clip_timestamps = clip_timestamps[self.initial_fps_frames:]
if len(clip) > 0:
interleave_clips.extend(list(clip.split(self.streaming_fps_frames)))
interleave_timestamps.extend(list(clip_timestamps.split(self.streaming_fps_frames)))
# 5. make conversation and send to model
for clip, timestamps in zip(interleave_clips, interleave_timestamps):
start_timestamp, stop_timestamp = timestamps[0].item(), timestamps[-1].item() + self.frame_time_interval
message = {
"role": "user",
"content": [
{"type": "text", "text": f'Time={start_timestamp:.1f}-{stop_timestamp:.1f}s'},
{"type": "video", "video": clip}
]
}
if not query and not state.get('query', None):
query = default_query
logger.warning(f'No query provided, use default_query={default_query}')
if query and state.get('query', None) != query:
message['content'].append({"type": "text", "text": query})
state['query'] = query
texts = self.processor.apply_chat_template([message], tokenize=False, add_generation_prompt=True, return_tensors='pt')
past_ids = state.get('past_ids', None)
if past_ids is not None:
texts = '<|im_end|>\n' + texts[self.system_prompt_offset:]
inputs = self.processor(
text=texts,
images=None,
videos=[clip],
return_tensors="pt",
return_attention_mask=False
)
inputs.to('cuda')
if past_ids is not None:
inputs['input_ids'] = torch.cat([past_ids, inputs.input_ids], dim=1)
if streaming_eos_base_threshold is not None:
logits_processor = [ThresholdLogitsProcessor(self.streaming_eos_token_id, streaming_eos_base_threshold, streaming_eos_threshold_step)]
else:
logits_processor = None
outputs = self.model.generate(
**inputs, past_key_values=state.get('past_key_values', None),
return_dict_in_generate=True, do_sample=do_sample,
repetition_penalty=repetition_penalty,
logits_processor=logits_processor,
)
state['past_key_values'] = outputs.past_key_values
state['past_ids'] = outputs.sequences[:, :-1]
yield (start_timestamp, stop_timestamp), self.processor.decode(outputs.sequences[0, inputs.input_ids.size(1):], skip_special_tokens=True), state
model_path = 'chenjoya/LiveCC-7B-Instruct'
video_path = "spacex_falcon9.mp4"
query = """Let's wait together!"""
infer = LiveCCDemoInfer(model_path=model_path)
state = {'video_path': video_path}
commentaries = []
t = 0
for t in range(31):
state['video_timestamp'] = t
for (start_t, stop_t), response, state in infer.live_cc(
query=query, state=state,
max_pixels = 512 * 28 * 28, repetition_penalty=1.05,
streaming_eos_base_threshold=0.0, streaming_eos_threshold_step=0
):
print(f'{start_t}s-{stop_t}s: {response}')
commentaries.append([start_t, stop_t, response])
if state.get('video_end', False):
break
t += 1
Limitations
- This model is starting from Qwen2-VL-7B-Base, so it may have limitations mentioned in https://huggingface.co/Qwen/Qwen2-VL-7B.
- This model is trained only with streaming frame-words paradigm, thus it may be only capable for real-time video commentary.
- The training ASR data is from YouTube CC, which has well-known low quality, so its formatting is not good (e.g. cannot output punctuation).
These limitations serve as ongoing directions for model optimization and improvement, and we are committed to continually enhancing the model's performance and scope of application.
Citation
If you find our work helpful, feel free to give us a cite.
@article{livecc,
author = {Joya Chen and Ziyun Zeng and Yiqi Lin and Wei Li and Zejun Ma and Mike Zheng Shou},
title = {LiveCC: Learning Video LLM with Streaming Speech Transcription at Scale},
journal = {arXiv preprint arXiv:2504.16030}
year = {2025},
}