File size: 2,091 Bytes
9b6e0aa | 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 |
import argparse
import json
from pathlib import Path
import soundfile as sf
MHB_ROOT = Path(__file__).resolve().parent.parent
def main():
parser = argparse.ArgumentParser()
parser.add_argument("test_dir", type=Path)
args = parser.parse_args()
test_dir = args.test_dir.expanduser().resolve()
assert test_dir.is_dir(), f"Not found: {test_dir}"
count = 0
with (MHB_ROOT / "meta_data.jsonl").open("r", encoding="utf-8") as f:
for line in f:
if not line.strip():
continue
record = json.loads(line)
if record["task_name"] != "ShipsEar":
continue
name = record["meta_data"]["ori_file_path"]
# Trawler 使用划分脚本生成的独立测试片段
if name.startswith("28__"):
name = Path(name).stem + "_test.wav"
matches = list(test_dir.glob(f"*{name}"))
assert len(matches) == 1, f"Expected one source for {name}, found {len(matches)}"
length = record["input_ts"]["length"]
sr = record["input_ts"]["fs"]
index = record["meta_data"]["seg_index"]
assert length == sr * 30
# 按 30 秒切分
with sf.SoundFile(matches[0]) as audio:
assert audio.samplerate == sr and audio.channels == 1, f"{matches[0], audio.samplerate, audio.channels}"
start = index * (sr * 15)
assert start + length <= audio.frames, f"Short recording: {matches[0]}"
audio.seek(start)
segment = audio.read(length, dtype="float32")
# 保存到 meta_data.jsonl 指定的路径
dst = MHB_ROOT / record["input_ts"]["path"]
assert dst.name == f"{record['id']}.wav"
dst.parent.mkdir(parents=True, exist_ok=True)
sf.write(dst, segment, sr, subtype="FLOAT")
count += 1
assert count == 84, f"Expected 84 segments, got {count}"
print(f"ShipsEar: created {count} WAV files")
if __name__ == "__main__":
main()
|