File size: 12,358 Bytes
ae8b600 e5e756a ae8b600 e5e756a ae8b600 cc826a1 e5e756a ae8b600 e5e756a ae8b600 cc826a1 ae8b600 e5e756a ae8b600 e5e756a cc826a1 e5e756a ae8b600 e5e756a ae8b600 e5e756a ae8b600 e5e756a cc826a1 e5e756a ae8b600 e5e756a | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 | import base64
import tempfile
from pathlib import Path
from fastapi import APIRouter, HTTPException, UploadFile, Query
from pydantic import BaseModel
from .core import get_video_audio_service
from plugins.audio.asr_drivers import get_asr_registry
from app.plugins.run_log import get_run_log_service, RunStatus
router = APIRouter()
plugin = None
def set_plugin_instance(plugin_instance):
global plugin
plugin = plugin_instance
class VideoBase64Request(BaseModel):
video_base64: str
suffix: str = ".mp4"
language: str | None = None
provider: str | None = None
@router.get("/status")
async def get_status():
if plugin is None:
return {
"name": "video",
"enabled": False,
"message": "插件未加载",
}
return plugin.get_status()
@router.post("/upload/transcribe")
async def transcribe_video_upload(
file: UploadFile,
language: str | None = None,
provider: str | None = None,
wait: bool = Query(False, description="是否同步等待转录完成"),
):
"""上传视频文件,提取音频并转录
默认异步返回 run_id,设置 wait=true 同步等待结果。
"""
if plugin is None or not plugin.enabled:
raise HTTPException(status_code=400, detail="插件未启用")
# 获取 ASR registry
registry = get_asr_registry()
provider_name = provider or registry.get_default_provider()
driver = registry.get_driver(provider_name)
if driver is None:
raise HTTPException(status_code=400, detail=f"未知的 ASR provider: {provider_name}")
# 检查 provider 可用性
availability = driver.check_availability()
if availability.value != "available":
raise HTTPException(
status_code=400,
detail=f"ASR provider {provider_name} 不可用: {availability.value}",
)
# 创建 run
run_service = get_run_log_service()
run = run_service.create_run("video")
# 写入上传事件
run_service.add_event(
run_id=run.run_id,
stage="upload",
message=f"视频文件已上传: {file.filename}",
detail=f"provider={provider_name}, language={language}",
)
# 保存上传文件到临时目录
content = await file.read()
suffix = "." + file.filename.rsplit(".", 1)[-1] if "." in file.filename else ".mp4"
with tempfile.NamedTemporaryFile(delete=False, suffix=suffix) as tmp_file:
tmp_file.write(content)
tmp_path = tmp_file.name
try:
# 写入音频提取开始事件
run_service.add_event(
run_id=run.run_id,
stage="extract_audio",
message="开始从视频提取音频",
)
# 提取音频
video_service = get_video_audio_service()
extraction = video_service.extract_audio(tmp_path)
if not extraction["success"]:
# 写入提取失败事件
run_service.add_event(
run_id=run.run_id,
stage="extract_audio",
message=f"音频提取失败: {extraction['error']}",
level="error",
)
run_service.finish_run(
run_id=run.run_id,
status=RunStatus.FAILED,
error=extraction["error"],
)
raise HTTPException(status_code=400, detail=extraction["error"])
# 写入音频提取完成事件
run_service.add_event(
run_id=run.run_id,
stage="extract_audio",
message="音频提取完成",
)
# 写入模型探测事件
run_service.add_event(
run_id=run.run_id,
stage="model_probe",
message=f"模型探测: {driver.model_id}",
detail=f"provider={provider_name}, status={availability.value}",
)
# 写入转录开始事件
run_service.add_event(
run_id=run.run_id,
stage="transcribe",
message="开始转录",
)
# 执行转录
audio_path = extraction["audio_path"]
result = driver.transcribe(audio_path, language)
# 清理提取的音频文件
Path(audio_path).unlink(missing_ok=True)
if result.success:
# 写入转录完成事件
run_service.add_event(
run_id=run.run_id,
stage="transcribe",
message=f"转录完成: {len(result.text)} 字符",
detail=f"timings={result.timings}",
)
# 完成 run
run_service.finish_run(
run_id=run.run_id,
status=RunStatus.SUCCEEDED,
result={
"text": result.text,
"segments": result.segments,
"provider": result.provider,
"model": result.model,
"timings": result.timings,
},
)
else:
# 写入转录失败事件
run_service.add_event(
run_id=run.run_id,
stage="transcribe",
message=f"转录失败: {result.error}",
level="error",
)
# 完成 run(失败)
run_service.finish_run(
run_id=run.run_id,
status=RunStatus.FAILED,
error=result.error,
)
# 如果同步等待,直接返回结果
if wait:
return {
"run_id": run.run_id,
"status": "succeeded" if result.success else "failed",
"provider": result.provider,
"model": result.model,
"text": result.text,
"segments": result.segments,
"timings": result.timings,
"error": result.error,
}
# 异步返回 run_id
return {
"run_id": run.run_id,
"status": "running",
"provider": provider_name,
"model": driver.model_id,
}
except HTTPException:
raise
except Exception as e:
# 写入异常事件
run_service.add_event(
run_id=run.run_id,
stage="transcribe",
message=f"转录异常: {str(e)}",
level="error",
)
run_service.finish_run(
run_id=run.run_id,
status=RunStatus.FAILED,
error=str(e),
)
raise HTTPException(status_code=500, detail=f"转录失败: {str(e)}")
finally:
# 清理临时文件
Path(tmp_path).unlink(missing_ok=True)
@router.post("/video-base64/transcribe")
async def transcribe_video_base64(request: VideoBase64Request):
"""Base64 视频转录(兼容旧接口)"""
if plugin is None or not plugin.enabled:
raise HTTPException(status_code=400, detail="插件未启用")
# 获取 ASR registry
registry = get_asr_registry()
provider_name = request.provider or registry.get_default_provider()
driver = registry.get_driver(provider_name)
if driver is None:
raise HTTPException(status_code=400, detail=f"未知的 ASR provider: {provider_name}")
# 检查 provider 可用性
availability = driver.check_availability()
if availability.value != "available":
raise HTTPException(
status_code=400,
detail=f"ASR provider {provider_name} 不可用: {availability.value}",
)
# 创建 run
run_service = get_run_log_service()
run = run_service.create_run("video")
# 写入上传事件
run_service.add_event(
run_id=run.run_id,
stage="upload",
message="Base64 视频已接收",
detail=f"provider={provider_name}, language={request.language}",
)
# 解码 Base64 并保存到临时文件
try:
video_bytes = base64.b64decode(request.video_base64)
except Exception as e:
run_service.finish_run(run_id=run.run_id, status=RunStatus.FAILED, error=f"Base64 解码失败: {e}")
raise HTTPException(status_code=400, detail=f"Base64 解码失败: {e}")
suffix = request.suffix if request.suffix.startswith(".") else f".{request.suffix}"
with tempfile.NamedTemporaryFile(delete=False, suffix=suffix) as tmp_file:
tmp_file.write(video_bytes)
tmp_path = tmp_file.name
try:
# 写入音频提取开始事件
run_service.add_event(
run_id=run.run_id,
stage="extract_audio",
message="开始从视频提取音频",
)
# 提取音频
video_service = get_video_audio_service()
extraction = video_service.extract_audio(tmp_path)
if not extraction["success"]:
# 写入提取失败事件
run_service.add_event(
run_id=run.run_id,
stage="extract_audio",
message=f"音频提取失败: {extraction['error']}",
level="error",
)
run_service.finish_run(
run_id=run.run_id,
status=RunStatus.FAILED,
error=extraction["error"],
)
return {
"run_id": run.run_id,
"status": "failed",
"error": extraction["error"],
}
# 写入音频提取完成事件
run_service.add_event(
run_id=run.run_id,
stage="extract_audio",
message="音频提取完成",
)
# 写入模型探测事件
run_service.add_event(
run_id=run.run_id,
stage="model_probe",
message=f"模型探测: {driver.model_id}",
detail=f"provider={provider_name}, status={availability.value}",
)
# 写入转录开始事件
run_service.add_event(
run_id=run.run_id,
stage="transcribe",
message="开始转录",
)
# 执行转录
audio_path = extraction["audio_path"]
result = driver.transcribe(audio_path, request.language)
# 清理提取的音频文件
Path(audio_path).unlink(missing_ok=True)
if result.success:
# 写入转录完成事件
run_service.add_event(
run_id=run.run_id,
stage="transcribe",
message=f"转录完成: {len(result.text)} 字符",
detail=f"timings={result.timings}",
)
# 完成 run
run_service.finish_run(
run_id=run.run_id,
status=RunStatus.SUCCEEDED,
result={
"text": result.text,
"segments": result.segments,
"provider": result.provider,
"model": result.model,
"timings": result.timings,
},
)
else:
# 写入转录失败事件
run_service.add_event(
run_id=run.run_id,
stage="transcribe",
message=f"转录失败: {result.error}",
level="error",
)
# 完成 run(失败)
run_service.finish_run(
run_id=run.run_id,
status=RunStatus.FAILED,
error=result.error,
)
return {
"run_id": run.run_id,
"status": "succeeded" if result.success else "failed",
"provider": result.provider,
"model": result.model,
"text": result.text,
"segments": result.segments,
"timings": result.timings,
"error": result.error,
}
except Exception as e:
# 写入异常事件
run_service.add_event(
run_id=run.run_id,
stage="transcribe",
message=f"转录异常: {str(e)}",
level="error",
)
run_service.finish_run(
run_id=run.run_id,
status=RunStatus.FAILED,
error=str(e),
)
raise HTTPException(status_code=500, detail=f"转录失败: {str(e)}")
finally:
# 清理临时文件
Path(tmp_path).unlink(missing_ok=True)
|