python .\faster_whisper_test_gen_srt.py (同じディレクトリのinput.mp4が読み込まれる) https://gist.github.com/taroyanaka/952513581d39065509bb39be3889b2c7

faster_whisper_test_gen_srt.py

######## whisperのモデルダウンロード ########
# from faster_whisper import WhisperModel
#
# model = WhisperModel("large-v3")
#
# print("モデルDL完了")
#############################################

from faster_whisper import WhisperModel

# =========================
# Whisper
# =========================

model = WhisperModel(
    "large-v3",
    device="cpu",
    compute_type="int8"
)

segments, info = model.transcribe(
    "input.mp4",
    language="ja"
)

# generator を全部読み切る
segments = list(segments)

print(f"検出言語: {info.language}")

for segment in segments:

    print(
        f"[{segment.start:.2f}s - "
        f"{segment.end:.2f}s] "
        f"{segment.text}"
    )

print("文字起こし完了")


# =========================
# SRT 時間変換
# =========================

def format_srt_time(seconds):

    hours = int(seconds // 3600)

    minutes = int(
        (seconds % 3600) // 60
    )

    secs = int(seconds % 60)

    milliseconds = int(
        (seconds - int(seconds)) * 1000
    )

    return (
        f"{hours:02}:"
        f"{minutes:02}:"
        f"{secs:02},"
        f"{milliseconds:03}"
    )


# =========================
# SRT 出力
# =========================

output_srt = "subtitle.srt"

with open(
    output_srt,
    "w",
    encoding="utf-8"
) as f:

    index = 1

    for segment in segments:

        text = segment.text.strip()

        if not text:
            continue

        start_time = format_srt_time(
            segment.start
        )

        end_time = format_srt_time(
            segment.end
        )

        f.write(f"{index}\n")

        f.write(
            f"{start_time} --> "
            f"{end_time}\n"
        )

        f.write(f"{text}\n\n")

        index += 1

print(
    f"SRTファイル出力完了: "
    f"{output_srt}"
)