Files
xiaoxia-saas/apps/worker/video_processing/speed_engine.py
T
xiaoxia 2081c72be6
CI/CD Pipeline / Unit Tests (push) Successful in 1m8s
CI/CD Pipeline / Frontend Lint (push) Successful in 1m18s
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 2m3s
CI/CD Pipeline / Integration Tests (push) Successful in 1m12s
CI/CD Pipeline / Build Production Runtime Images (push) Successful in 5m31s
CI/CD Pipeline / Deploy Production (push) Failing after 11s
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
feat: 视频调速引擎(快进/慢放) (#294)
2026-07-14 11:36:24 +08:00

168 lines
5.2 KiB
Python
Executable File
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
"""视频调速引擎 — 基于 FFmpeg setpts + atempo 的速度调整能力。
支持:
- 0.25x ~ 4x 变速范围
- 视频调速(setpts
- 音频调速(atempo,多级串联处理超范围值)
- 音调修正(pitch_correct,默认开启)
- 边界自动钳制,不阻断渲染
"""
from dataclasses import dataclass
from typing import Optional
# ─── 常量 ───────────────────────────────────────────────
MIN_SPEED = 0.25
MAX_SPEED = 4.0
DEFAULT_SPEED = 1.0
# atempo 单级有效范围
_ATEMPO_MIN = 0.5
_ATEMPO_MAX = 2.0
@dataclass
class SpeedConfig:
"""调速配置。
Attributes:
speed: 播放速度,0.25~4.01.0 为原速
pitch_correct: 是否保持音调(默认 True,用 atempo 时间拉伸算法)
"""
speed: float = DEFAULT_SPEED
pitch_correct: bool = True
@classmethod
def parse(cls, data: Optional[dict]) -> "SpeedConfig":
"""从 dict 解析配置,无效值回退到默认。"""
if not data or not isinstance(data, dict):
return cls()
speed = data.get("speed", DEFAULT_SPEED)
if not isinstance(speed, (int, float)):
speed = DEFAULT_SPEED
pitch_correct = data.get("pitch_correct", True)
if not isinstance(pitch_correct, bool):
pitch_correct = True
config = cls(speed=float(speed), pitch_correct=pitch_correct)
config.clamp()
return config
def clamp(self) -> None:
"""将速度钳制到合法范围。"""
if self.speed <= 0:
self.speed = DEFAULT_SPEED
elif self.speed < MIN_SPEED:
self.speed = MIN_SPEED
elif self.speed > MAX_SPEED:
self.speed = MAX_SPEED
@property
def is_original(self) -> bool:
"""是否原速(无需调速)。"""
return abs(self.speed - 1.0) < 1e-6
class SpeedEngine:
"""调速引擎 — 生成 FFmpeg 调速滤镜链。
用法:
engine = SpeedEngine()
video_filter = engine.build_video_filter(config)
audio_filter = engine.build_audio_filter(config)
new_duration = engine.adjust_duration(duration, config)
"""
def build_video_filter(self, config: SpeedConfig) -> str:
"""生成视频调速滤镜字符串。
返回 setpts 滤镜表达式,原速时返回空字符串。
"""
if config.is_original:
return ""
# setpts=PTS/speed — speed>1 加速,speed<1 减速
return f"setpts=PTS/{config.speed:.4f}"
def build_audio_filter(self, config: SpeedConfig) -> str:
"""生成音频调速滤镜字符串。
atempo 单级范围 0.5~2.0,超出范围时自动多级串联:
- 0.25x → atempo=0.5,atempo=0.5
- 4x → atempo=2.0,atempo=2.0
- 0.3x → atempo=0.5,atempo=0.6
- 3x → atempo=2.0,atempo=1.5
原速时返回空字符串。
"""
if config.is_original:
return ""
speed = config.speed
stages: list[float] = self._split_atempo_stages(speed)
return ",".join(f"atempo={s:.4f}" for s in stages)
@staticmethod
def _split_atempo_stages(speed: float) -> list[float]:
"""将速度拆分为多级 atempo 串联,每级都在 [0.5, 2.0] 范围内。"""
if _ATEMPO_MIN <= speed <= _ATEMPO_MAX:
return [speed]
stages: list[float] = []
remaining = speed
# 加速场景(speed > 2.0
if speed > _ATEMPO_MAX:
while remaining > _ATEMPO_MAX:
stages.append(_ATEMPO_MAX)
remaining /= _ATEMPO_MAX
stages.append(remaining)
# 减速场景(speed < 0.5
else:
while remaining < _ATEMPO_MIN:
stages.append(_ATEMPO_MIN)
remaining /= _ATEMPO_MIN
stages.append(remaining)
return stages
def adjust_duration(self, original_duration: float, config: SpeedConfig) -> float:
"""计算调速后的时长。
加速 → 时长变短;减速 → 时长变长。
"""
if config.is_original or original_duration <= 0:
return original_duration
return original_duration / config.speed
def build_clip_speed_filter(
self,
speed: float,
pitch_correct: bool = True,
) -> tuple[str, str, SpeedConfig]:
"""便捷方法:从单一 speed 值生成视频+音频滤镜。
返回 (video_filter, audio_filter, config)。
"""
config = SpeedConfig(speed=speed, pitch_correct=pitch_correct)
config.clamp()
return (
self.build_video_filter(config),
self.build_audio_filter(config),
config,
)
@staticmethod
def resolve_clip_speed(
clip_config: dict,
global_speed: float = DEFAULT_SPEED,
) -> float:
"""从 clip config 中解析 playback_speed0 或缺失则使用全局速度。"""
speed = clip_config.get("playback_speed", 0) if clip_config else 0
if not isinstance(speed, (int, float)) or speed <= 0:
return global_speed
return float(speed)