Spaces:
Running
on
Zero
Running
on
Zero
File size: 11,858 Bytes
a602628 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 |
"""
Timeline Manager - Manages timeline-based audio generation and blending
"""
import numpy as np
import torch
import torchaudio
from pathlib import Path
from typing import Optional, Dict, List, Any
import json
import logging
from datetime import datetime
import matplotlib.pyplot as plt
import matplotlib
matplotlib.use('Agg') # Non-interactive backend
logger = logging.getLogger(__name__)
class TimelineManager:
"""Manages audio timeline with seamless blending."""
def __init__(self, config: Dict[str, Any]):
"""
Initialize timeline manager.
Args:
config: Configuration dictionary
"""
self.config = config
self.sample_rate = config.get("sample_rate", 44100)
self.timelines = {} # Store active timelines
self.timeline_dir = Path(config.get("timeline_dir", "timelines"))
self.timeline_dir.mkdir(exist_ok=True)
def create_timeline(self) -> str:
"""
Create new timeline.
Returns:
Timeline ID
"""
timeline_id = datetime.now().strftime("%Y%m%d_%H%M%S_%f")
self.timelines[timeline_id] = {
"id": timeline_id,
"clips": [],
"audio": None,
"metadata": [],
"created_at": datetime.now().isoformat()
}
logger.info(f"Created timeline: {timeline_id}")
return timeline_id
def add_clip(
self,
timeline_id: Optional[str],
clip_path: str,
metadata: Dict[str, Any]
) -> str:
"""
Add clip to timeline.
Args:
timeline_id: Timeline ID (creates new if None)
clip_path: Path to audio clip
metadata: Clip metadata
Returns:
Timeline ID
"""
try:
# Create timeline if doesn't exist
if timeline_id is None or timeline_id not in self.timelines:
timeline_id = self.create_timeline()
timeline = self.timelines[timeline_id]
# Load clip
clip_audio, sr = torchaudio.load(clip_path)
# Resample if needed
if sr != self.sample_rate:
resampler = torchaudio.transforms.Resample(sr, self.sample_rate)
clip_audio = resampler(clip_audio)
# Convert to numpy
clip_np = clip_audio.numpy()
# Add to timeline
if timeline["audio"] is None:
# First clip
timeline["audio"] = clip_np
else:
# Concatenate with existing audio
timeline["audio"] = np.concatenate([timeline["audio"], clip_np], axis=1)
# Store metadata
clip_info = {
"index": len(timeline["clips"]),
"path": clip_path,
"duration": clip_np.shape[1] / self.sample_rate,
"start_time": self.get_duration(timeline_id) - (clip_np.shape[1] / self.sample_rate),
"metadata": metadata
}
timeline["clips"].append(clip_info)
logger.info(f"Added clip to timeline {timeline_id}: {clip_info['duration']:.2f}s")
return timeline_id
except Exception as e:
logger.error(f"Failed to add clip: {e}")
raise
def get_context(
self,
timeline_id: Optional[str],
context_length: int
) -> Optional[np.ndarray]:
"""
Get context audio from timeline.
Args:
timeline_id: Timeline ID
context_length: Length in seconds to retrieve
Returns:
Context audio as numpy array or None
"""
if timeline_id is None or timeline_id not in self.timelines:
return None
timeline = self.timelines[timeline_id]
if timeline["audio"] is None:
return None
# Calculate number of samples
context_samples = int(context_length * self.sample_rate)
# Get last N samples
audio = timeline["audio"]
if audio.shape[1] <= context_samples:
return audio
return audio[:, -context_samples:]
def get_last_clip(self, timeline_id: Optional[str]) -> Optional[np.ndarray]:
"""Get the last clip from timeline."""
if timeline_id is None or timeline_id not in self.timelines:
return None
timeline = self.timelines[timeline_id]
if not timeline["clips"]:
return None
last_clip = timeline["clips"][-1]
audio, _ = torchaudio.load(last_clip["path"])
return audio.numpy()
def export_timeline(self, timeline_id: str) -> str:
"""
Export full timeline audio.
Args:
timeline_id: Timeline ID
Returns:
Path to exported audio file
"""
if timeline_id not in self.timelines:
raise ValueError(f"Timeline not found: {timeline_id}")
timeline = self.timelines[timeline_id]
if timeline["audio"] is None:
raise ValueError("Timeline is empty")
# Save to file
output_path = self.timeline_dir / f"timeline_{timeline_id}.wav"
audio_tensor = torch.from_numpy(timeline["audio"]).float()
torchaudio.save(
str(output_path),
audio_tensor,
self.sample_rate,
encoding="PCM_S",
bits_per_sample=16
)
logger.info(f"Exported timeline to {output_path}")
return str(output_path)
def visualize_timeline(self, timeline_id: str) -> str:
"""
Create visualization of timeline.
Args:
timeline_id: Timeline ID
Returns:
Path to visualization image
"""
if timeline_id not in self.timelines:
raise ValueError(f"Timeline not found: {timeline_id}")
timeline = self.timelines[timeline_id]
if not timeline["clips"]:
# Create empty visualization
fig, ax = plt.subplots(figsize=(12, 4))
ax.text(0.5, 0.5, "No clips yet", ha='center', va='center')
ax.set_xlim(0, 1)
ax.set_ylim(0, 1)
else:
# Create timeline visualization
fig, ax = plt.subplots(figsize=(12, 4))
total_duration = self.get_duration(timeline_id)
# Draw each clip
for clip in timeline["clips"]:
start = clip["start_time"]
duration = clip["duration"]
# Draw clip rectangle
rect = plt.Rectangle(
(start, 0.3),
duration,
0.4,
facecolor='steelblue',
edgecolor='black',
linewidth=1
)
ax.add_patch(rect)
# Add clip label
ax.text(
start + duration/2,
0.5,
f"Clip {clip['index'] + 1}",
ha='center',
va='center',
fontsize=8,
color='white',
weight='bold'
)
ax.set_xlim(0, max(total_duration, 1))
ax.set_ylim(0, 1)
ax.set_xlabel('Time (seconds)', fontsize=10)
ax.set_title(f'Timeline: {len(timeline["clips"])} clips, {total_duration:.1f}s total', fontsize=12)
ax.set_yticks([])
ax.grid(True, axis='x', alpha=0.3)
# Save visualization
viz_path = self.timeline_dir / f"timeline_{timeline_id}_viz.png"
plt.tight_layout()
plt.savefig(viz_path, dpi=100, bbox_inches='tight')
plt.close()
return str(viz_path)
def get_duration(self, timeline_id: str) -> float:
"""Get total duration of timeline in seconds."""
if timeline_id not in self.timelines:
return 0.0
timeline = self.timelines[timeline_id]
if timeline["audio"] is None:
return 0.0
return timeline["audio"].shape[1] / self.sample_rate
def inpaint_region(
self,
timeline_id: str,
start_time: float,
end_time: float,
new_prompt: str
) -> str:
"""
Inpaint specific region in timeline.
Args:
timeline_id: Timeline ID
start_time: Start time in seconds
end_time: End time in seconds
new_prompt: Prompt for new content
Returns:
Path to updated timeline audio
"""
if timeline_id not in self.timelines:
raise ValueError(f"Timeline not found: {timeline_id}")
# This would integrate with ACE-Step engine for actual inpainting
# For now, this is a placeholder
logger.info(f"Inpainting {start_time:.1f}s-{end_time:.1f}s in timeline {timeline_id}")
# Export current state
return self.export_timeline(timeline_id)
def delete_timeline(self, timeline_id: str):
"""Delete timeline and associated files."""
if timeline_id in self.timelines:
del self.timelines[timeline_id]
logger.info(f"Deleted timeline: {timeline_id}")
def save_timeline_state(self, timeline_id: str):
"""Save timeline state to disk."""
if timeline_id not in self.timelines:
return
timeline = self.timelines[timeline_id]
# Save metadata
metadata_path = self.timeline_dir / f"timeline_{timeline_id}_metadata.json"
metadata = {
"id": timeline["id"],
"clips": timeline["clips"],
"created_at": timeline["created_at"],
"duration": self.get_duration(timeline_id)
}
with open(metadata_path, 'w') as f:
json.dump(metadata, f, indent=2)
# Export audio
if timeline["audio"] is not None:
self.export_timeline(timeline_id)
logger.info(f"Saved timeline state: {timeline_id}")
def load_timeline_state(self, timeline_id: str) -> bool:
"""Load timeline state from disk."""
metadata_path = self.timeline_dir / f"timeline_{timeline_id}_metadata.json"
audio_path = self.timeline_dir / f"timeline_{timeline_id}.wav"
if not metadata_path.exists():
return False
try:
# Load metadata
with open(metadata_path, 'r') as f:
metadata = json.load(f)
# Load audio if exists
audio = None
if audio_path.exists():
audio_tensor, _ = torchaudio.load(str(audio_path))
audio = audio_tensor.numpy()
# Restore timeline
self.timelines[timeline_id] = {
"id": timeline_id,
"clips": metadata["clips"],
"audio": audio,
"metadata": [],
"created_at": metadata["created_at"]
}
logger.info(f"Loaded timeline state: {timeline_id}")
return True
except Exception as e:
logger.error(f"Failed to load timeline: {e}")
return False
|