apps/audio_cloning/vallex/examples.py

_prompts_dir = "apps/audio_cloning/vallex/prompts"

infer_from_audio_examples = [
    [
        "私のクローンに騙されないでください。",
        "日本語",
        "no-accent",
        f"{_prompts_dir}/ja-okuwaki.wav",
        None,
        "こんにちは、私の名前はオクワキヨウスケです。",
    ],
    [
        "ぼくのクローンに騙されないでくれなのだ。",
        "日本語",
        "no-accent",
        f"{_prompts_dir}/ja-zundamon.wav",
        None,
        "はじめまして、ずんだもんなのだ",
    ],
    [
        "私のクローンに騙されないでください。",
        "日本語",
        "no-accent",
        f"{_prompts_dir}/ja-okuwaki-long.wav",
        None,
        "こんにちは、私の名前はオクワキヨウスケです。これは音声クローニング用のサンプルです。",
    ],
    [
        "私の声を真似するのはそんなに面白いですか？",
        "日本語",
        "no-accent",
        f"{_prompts_dir}/ja-2.ogg",
        None,
        "初めまして、朝武よしのです。",
    ],
    [
        "This is how this machine has taken my voice.",
        "English",
        "no-accent",
        f"{_prompts_dir}/en-2.wav",
        None,
        "Wow, look at that! That's no ordinary Teddy bear!",
    ],
    [
        "我喜欢抽电子烟，尤其是锐刻五代。",
        "中文",
        "no-accent",
        f"{_prompts_dir}/zh-1.wav",
        None,
        "今天我很荣幸，",
    ],
    [
        "你可以听得出来我有多困。",
        "中文",
        "no-accent",
        f"{_prompts_dir}/en-1.wav",
        None,
        "",
    ],
    [
        "この文は、クロスリンガル合成の例です。",
        "日本語",
        "no-accent",
        f"{_prompts_dir}/zh-2.wav",
        None,
        "",
    ],
    [
        "Actually, I can't speak English, but this machine helped me do it.",
        "English",
        "no-accent",
        f"{_prompts_dir}/ja-1.wav",
        None,
        "",
    ],
]

make_npz_prompt_examples = [
    [
        "Gem-trader",
        f"{_prompts_dir}/en-2.wav",
        None,
        "Wow, look at that! That's no ordinary Teddy bear!",
    ],
    ["Ding Zhen", f"{_prompts_dir}/zh-1.wav", None, "今天我很荣幸，"],
    ["Yoshino", f"{_prompts_dir}/ja-2.ogg", None, "初めまして、朝武よしのです。"],
    ["Sleepy-woman", f"{_prompts_dir}/en-1.wav", None, ""],
    ["Yae", f"{_prompts_dir}/zh-2.wav", None, ""],
    ["Cafe", f"{_prompts_dir}/ja-1.wav", None, ""],
]

infer_from_prompt_examples = [
    [
        "A prompt contains voice, prosody and emotion information of a certain speaker.",
        "English",
        "no-accent",
        f"{_prompts_dir}/vctk_1",
        None,
    ],
    [
        "This prompt is made with an audio of three seconds.",
        "English",
        "no-accent",
        f"{_prompts_dir}/librispeech_1",
        None,
    ],
    ["This prompt is made with Chinese speech", "English", "no-accent", "seel", None],
]