Gemini API で動画から多言語字幕（SRT/VTT）を生成する実装パターン

from google import genai
from google.genai import types
import time
 
client = genai.Client(api_key="YOUR_API_KEY")
 
# 動画ファイルをアップロード
video_file = client.files.upload(file="meeting.mp4")
print(f"Uploaded: {video_file.name}, state: {video_file.state}")
 
# state が ACTIVE になるまでポーリング
while video_file.state == "PROCESSING":
    time.sleep(2)
    video_file = client.files.get(name=video_file.name)
print(f"Ready: {video_file.state}")  # 期待値: ACTIVE

config = types.GenerateContentConfig(
    response_mime_type="application/json",
    response_schema={
        "type": "ARRAY",
        "items": {
            "type": "OBJECT",
            "properties": {
                "start": {"type": "STRING", "description": "MM:SS.mmm format"},
                "end": {"type": "STRING", "description": "MM:SS.mmm format"},
                "ja": {"type": "STRING"},
                "en": {"type": "STRING"},
            },
            "required": ["start", "end", "ja", "en"]
        }
    }
)
 
prompt = """この動画の話者音声を聞き取り、字幕データを生成してください。
要件:
- タイムスタンプは MM:SS.mmm 形式（例: 01:23.450）
- 1字幕は最大35文字、2行まで（改行は \\n で表現）
- ja は日本語の話者発話そのまま、en は自然な英語訳
- 沈黙・BGMのみの区間は出力しない
"""
 
response = client.models.generate_content(
    model="gemini-2.5-pro",
    contents=[video_file, prompt],
    config=config,
)
 
import json
captions = json.loads(response.text)
print(f"生成字幕数: {len(captions)}")

def to_srt(captions: list[dict]) -> str:
    """captions は [{start, end, ja, en}, ...] 形式"""
    lines = []
    for i, c in enumerate(captions, 1):
        start = c["start"].replace(".", ",")  # SRT は ms を , で区切る
        end = c["end"].replace(".", ",")
        # MM:SS,mmm を HH:MM:SS,mmm に揃える
        start = f"00:{start}" if start.count(":") == 1 else start
        end = f"00:{end}" if end.count(":") == 1 else end
        lines.append(f"{i}\n{start} --> {end}\n{c['ja']}\n")
    return "\n".join(lines)
 
def to_vtt(captions: list[dict], lang: str = "ja") -> str:
    """VTT はヘッダ付き、ms は . 区切り"""
    body = []
    for c in captions:
        start = f"00:{c['start']}" if c['start'].count(":") == 1 else c['start']
        end = f"00:{c['end']}" if c['end'].count(":") == 1 else c['end']
        body.append(f"{start} --> {end}\n{c[lang]}")
    return "WEBVTT\n\n" + "\n\n".join(body)
 
with open("output_ja.srt", "w") as f:
    f.write(to_srt(captions))
with open("output_en.vtt", "w") as f:
    f.write(to_vtt(captions, lang="en"))

# まずは Flash で叩き台を生成
response = client.models.generate_content(model="gemini-2.5-flash", contents=[video_file, prompt], config=config)
# ノイズが多い素材は Pro で再生成
response = client.models.generate_content(model="gemini-2.5-pro", contents=[video_file, prompt], config=config)

def qa_captions(captions: list[dict]) -> list[str]:
    issues = []
    def to_seconds(ts: str) -> float:
        # MM:SS.mmm -> 秒
        m, rest = ts.split(":", 1)
        return int(m) * 60 + float(rest)
    prev_end = 0.0
    for i, c in enumerate(captions):
        s, e = to_seconds(c["start"]), to_seconds(c["end"])
        cps = len(c["ja"]) / max(e - s, 0.1)
        if cps > 17:
            issues.append(f"cue {i}: 速すぎ ({cps:.1f} cps)")
        if s < prev_end:
            issues.append(f"cue {i}: 前の字幕と重複")
        if s - prev_end > 10:
            issues.append(f"cue {i}: 直前と {s - prev_end:.1f}s の空白")
        prev_end = e
    return issues
 
for problem in qa_captions(captions):
    print(problem)