Use the audiox package to perform audio generation programmatically. This example demonstrates a Video-to-Music (V2M) generation task.
Workflow
- Load Model: Use
get_pretrained_model(model_name) to load the model and its configuration. - Prepare Inputs:
- Read video using
read_video. - Load audio using
load_and_process_audio (or use a zero tensor if no audio prompt is provided). - For
AudioX-MAF or AudioX-MAF-MMDiT models, encode video with encode_video_with_synchformer.
- Generate: Call
generate_diffusion_cond with the model and a conditioning list containing the prompts. - Post-process: Rescale and convert the output tensor to
int16 before saving with torchaudio.
import torch
import torchaudio
from einops import rearrange
from audiox import get_pretrained_model
from audiox.inference.generation import generate_diffusion_cond
from audiox.data.utils import read_video, merge_video_audio, load_and_process_audio, encode_video_with_synchformer
import os
device = "cuda" if torch.cuda.is_available() else "cpu"
# Load pretrained model
model_name = "HKUSTAudio/AudioX-MAF"
model, model_config = get_pretrained_model(model_name)
sample_rate = model_config["sample_rate"]
sample_size = model_config["sample_size"]
target_fps = model_config["video_fps"]
seconds_start = 0
seconds_total = 10
model = model.to(device)
# Example: Video-to-Music generation
video_path = "example/V2M_sample-1.mp4"
text_prompt = "Generate music for the video"
audio_path = None
# Prepare inputs
video_tensor = read_video(video_path, seek_time=seconds_start, duration=seconds_total, target_fps=target_fps)
if audio_path:
audio_tensor = load_and_process_audio(audio_path, sample_rate, seconds_start, seconds_total)
else:
# Use zero tensor when no audio is provided
audio_tensor = torch.zeros((2, int(sample_rate * seconds_total)))
# For AudioX-MAF and AudioX-MAF-MMDiT: encode video with synchformer
video_sync_frames = None
if "MAF" in model_name:
video_sync_frames = encode_video_with_synchformer(
video_path, model_name, seconds_start, seconds_total, device
)
# Create conditioning
conditioning = [{
"video_prompt": {"video_tensors": video_tensor.unsqueeze(0), "video_sync_frames": video_sync_frames},
"text_prompt": text_prompt,
"audio_prompt": audio_tensor.unsqueeze(0),
"seconds_start": seconds_start,
"seconds_total": seconds_total
}]
# Generate audio
output = generate_diffusion_cond(
model,
steps=250,
cfg_scale=7,
conditioning=conditioning,
sample_size=sample_size,
sigma_min=0.3,
sigma_max=500,
sampler_type="dpmpp-3m-sde",
device=device
)
# Post-process audio
output = rearrange(output, "b d n -> d (b n)")
output = output.to(torch.float32).div(torch.max(torch.abs(output))).clamp(-1, 1).mul(32767).to(torch.int16).cpu()
torchaudio.save("output.wav", output, sample_rate)
# Merge with video (optional)
if video_path is not None and os.path.exists(video_path):
merge_video_audio(video_path, "output.wav", "output.mp4", seconds_start, seconds_total)
import torch
import torchaudio
from einops import rearrange
from audiox import get_pretrained_model
from audiox.inference.generation import generate_diffusion_cond
from audiox.data.utils import read_video, merge_video_audio, load_and_process_audio, encode_video_with_synchformer
import os
device = "cuda" if torch.cuda.is_available() else "cpu"
# Load pretrained model
# Choose one: "HKUSTAudio/AudioX", "HKUSTAudio/AudioX-MAF", or "HKUSTAudio/AudioX-MAF-MMDiT"
model_name = "HKUSTAudio/AudioX-MAF"
model, model_config = get_pretrained_model(model_name)
sample_rate = model_config["sample_rate"]
sample_size = model_config["sample_size"]
target_fps = model_config["video_fps"]
seconds_start = 0
seconds_total = 10
model = model.to(device)
# Example: Video-to-Music generation
video_path = "example/V2M_sample-1.mp4"
text_prompt = "Generate music for the video"
audio_path = None
# Prepare inputs
video_tensor = read_video(video_path, seek_time=seconds_start, duration=seconds_total, target_fps=target_fps)
if audio_path:
audio_tensor = load_and_process_audio(audio_path, sample_rate, seconds_start, seconds_total)
else:
# Use zero tensor when no audio is provided (following gradio implementation)
audio_tensor = torch.zeros((2, int(sample_rate * seconds_total)))
# For AudioX-MAF and AudioX-MAF-MMDiT: encode video with synchformer
video_sync_frames = None
if "MAF" in model_name:
video_sync_frames = encode_video_with_synchformer(
video_path, model_name, seconds_start, seconds_total, device
)
# Create conditioning (always include audio_prompt, using zero tensor if no audio)
conditioning = [{
"video_prompt": {"video_tensors": video_tensor.unsqueeze(0), "video_sync_frames": video_sync_frames},
"text_prompt": text_prompt,
"audio_prompt": audio_tensor.unsqueeze(0),
"seconds_start": seconds_start,
"seconds_total": seconds_total
}]
# Generate audio
output = generate_diffusion_cond(
model,
steps=250,
cfg_scale=7,
conditioning=conditioning,
sample_size=sample_size,
sigma_min=0.3,
sigma_max=500,
sampler_type="dpmpp-3m-sde",
device=device
)
# Post-process audio
output = rearrange(output, "b d n -> d (b n)")
output = output.to(torch.float32).div(torch.max(torch.abs(output))).clamp(-1, 1).mul(32767).to(torch.int16).cpu()
torchaudio.save("output.wav", output, sample_rate)
# Merge with video (optional)
if video_path is not None and os.path.exists(video_path):
merge_video_audio(video_path, "output.wav", "output.mp4", seconds_start, seconds_total)
