# -*- coding: utf-8 -*- """开曲轻检:只切前奏窗做 ASR 衬词统计(比全曲 QC 快)。 用法(qc_python): python intro_check_worker.py '{"file":"...","bpm":104,"intro_bars":8,"lang":"zh"}' """ import json import os import sys import traceback HERE = os.path.dirname(os.path.abspath(__file__)) if HERE not in sys.path: sys.path.insert(0, HERE) def main(job): import numpy as np import qc_worker as q path = job["file"] bpm = float(job.get("bpm") or 104) bars = int(job.get("intro_bars") or 8) lang = job.get("lang") or "zh" intro_sec = bars * 60.0 / bpm * 4.0 # 与全量 qc 同口径:对整段 vocals 做 ASR,再按时间窗统计。 # 只切前奏短窗喂 Whisper 会严重少计衬词(实测只吐 1 个「嗯」却 PASS)。 v, vsr = q.vocals_of(path, job.get("cache_dir")) a = q.asr(v, vsr, lang) segs = a.get("segments") or [] stats = q.intro_filler_stats( segs, intro_sec, lang, voc=v, vsr=vsr) hard = q.HARD ivr = float(stats.get("intro_voc_rms") or 0) ifn = int(stats["intro_filler_n"]) ifr = float(stats["intro_filler_ratio"]) # 有人声能量时降低字数门槛:短窗幻觉「单个嗯」也要拦 eff_min = hard["intro_filler_min_chars"] if ivr >= hard["intro_voc_rms_min"]: eff_min = min(eff_min, 2) fail = (ifn >= eff_min and ifr >= hard["intro_filler_ratio"] and ivr >= hard["intro_voc_rms_min"]) # 前 2s 真唱字(content words)也算开曲违规 early = 0 for s in segs: if float(s.get("t0", 0)) >= 2.0: continue early += len(q.content_words(s.get("text") or "", lang)) if early > 0: fail = True return dict( ok=not fail, fail=fail, intro_sec=round(intro_sec, 2), early_content=early, asr_head=(a.get("text") or "")[:160], eff_min=eff_min, **stats, ) if __name__ == "__main__": try: job = json.loads(sys.argv[1]) out = main(job) except Exception as e: out = dict(error=str(e), type=type(e).__name__, trace=traceback.format_exc()[-800:]) print("###JSON###") print(json.dumps(out, ensure_ascii=False))