Chiloh

Chiloh Wei

一个尝试理解世界,理解自己的人
jike
twitter

05-當下不雜

image

封面圖拍攝於 21 年的 10 月 06 日,是我第一次進去夜場。和朋友要了個卡座,周邊形形色色,我在其中格格不入。


這周去了南昌出差,每次出差都滿是碳水和風土人情的款待。我話比較少,往往在聚餐的時候是一個觀察者,每一次都能發覺許多有意思的人,和有意思的點。真好呀~

這周在飛機上粗讀了這本書,非常好的一本書。看似在講 “創意”,實質卻是大道萬千,殊途同歸。分享下我最喜歡的這段話:

我們每一個人內心深處似乎都具有一個個人創意泉源。同時,存在一種更廣大、超越個人、屬於全人類的共同泉源,裡面儲存著各種原始、深奧的集體智慧。這個龐大泉源或許在我們體內,或許我們有管道可以通到它。

回來後,我在朋友圈分享了自己的感受,只有寥寥的 3 行文字:

image

看見 Vercel 也推出了自己的生成式 UI 系統,輸入文本,就可以繪製一個簡單的 UI 界面。與之前分享過的 screenshot-to-code 比較類似。

這周自己測試了下,用這個爬蟲框架,來爬取自己博客的內容,並將其餵給在訓練的自己的分身,效果還不錯。

感受#

這周在 AI 上的實踐,是基於開源項目:tts,借助其提供的方案,花了不到 2 分鐘,在 cloudflare 上就部署好了這個用 Azure tts 來做語音合成的網站:語音合成

image

整個方案非常簡單,在 cloudflare 上新建一個 workers,寫入以下代碼就可以實現:

addEventListener('fetch', event => {
    event.respondWith(handleRequest(event.request))
})


function generateUUID() {
    let uuid = 'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'.replace(/[x]/g, function (c) {
        let r = Math.random() * 16 | 0,
            v = c === 'x' ? r : (r & 0x3 | 0x8);
        return v.toString(16);
    });
    return uuid;
}

const API_URL = "https://southeastasia.api.speech.microsoft.com/accfreetrial/texttospeech/acc/v3.0-beta1/vcg/speak";
const DEFAULT_HEADERS = {
    authority: "southeastasia.api.speech.microsoft.com",
    accept: "*/*",
    "accept-language": "zh-CN,zh;q=0.9",
    customvoiceconnectionid: generateUUID(),
    origin: "https://speech.microsoft.com",
    "sec-ch-ua":
        '"Google Chrome";v="111", "Not(A:Brand";v="8", "Chromium";v="111"',
    "sec-ch-ua-mobile": "?0",
    "sec-ch-ua-platform": '"Windows"',
    "sec-fetch-dest": "empty",
    "sec-fetch-mode": "cors",
    "sec-fetch-site": "same-site",
    "user-agent":
        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/111.0.0.0 Safari/537.36",
    "content-type": "application/json",
};

const speechApi = async (ssml) => {
    const data = JSON.stringify({
        ssml,
        ttsAudioFormat: "audio-24khz-160kbitrate-mono-mp3",
        offsetInPlainText: 0,
        properties: {
            SpeakTriggerSource: "AccTuningPagePlayButton",
        },
    });

    try {
        const response = await fetch(API_URL, {
            method: "POST",
            responseType: "arraybuffer",
            headers: DEFAULT_HEADERS,
            body: data
        });

        if (!response.ok) {
            throw new Error(`Request failed with status ${response.status}`);
        }

        return response.arrayBuffer();
    } catch (error) {
        console.error("Error during API request:", error);
        throw error;
    }
};

const handleRequest = async (request) => {
    // 解析請求 URL
    const url = new URL(request.url);

    const clientIP = request.headers.get("CF-Connecting-IP")

    if (url.pathname == "/") {
        const html = await fetch("https://files.chiloh.net/azure-tts.html")

        const page = await html.text()
        return new Response(page, {
            headers: {
                "content-type": "text/html;charset=UTF-8",
                "Access-Control-Allow-Origin": "*",
                "Access-Control-Allow-Credentials": "true",
                "Access-Control-Allow-Headers": "*",
                "Access-Control-Allow-Methods": "*",
                "ip": `Access cloudflare's ip:${clientIP}`
            },
        })
    } else if (url.pathname == "/audio") {
        // 解析查詢參數
        const params = new URLSearchParams(url.search);
        // 獲取查詢參數中的文本
        const text = params.get("text");
        // 獲取查詢參數中的語速
        const rate = params.get("rate");
        // 獲取查詢參數中的音高
        const pitch = params.get("pitch");
        // 獲取查詢參數中的音色
        const voice = params.get("voice");
        // 獲取查詢參數中的音色風格
        const voiceStyle = params.get("voiceStyle");
        const ssml = `<speak xmlns="http://www.w3.org/2001/10/synthesis" xmlns:mstts="http://www.w3.org/2001/mstts" xmlns:emo="http://www.w3.org/2009/10/emotionml" version="1.0" xml:lang="en-US">
    <voice name="${voice}">
    <mstts:express-as style="${voiceStyle}">
        <prosody rate="${rate}%" pitch="${pitch}%">
        ${text}
       </prosody>
        </mstts:express-as>
    </voice>
    </speak>`;

        const audio = await speechApi(ssml);
        const nowtime = new Date().getTime();
        return new Response(audio, {
            headers: {
                "Content-Type": "audio/mpeg",
                "Content-Disposition": `attachment; filename=${nowtime}.mp3`,
            },
        });
    } else if (url.pathname == "/legado") {
        const origin = url.origin
        const params = new URLSearchParams(url.search);
        // 獲取查詢參數中的文本
        // const text = params.get("text");
        // 獲取查詢參數中的語速
        const rate = params.get("rate");
        // 獲取查詢參數中的音高
        const pitch = params.get("pitch");
        // 獲取查詢參數中的音色
        const voice = params.get("voice");
        // 獲取查詢參數中的音色風格
        const voiceStyle = params.get("voiceStyle");

        const dataJson = {
            "concurrentRate": "",//並發率
            "contentType": "audio/mpeg",
            "header": "",
            "id": Date.now(),
            "lastUpdateTime": Date.now(),
            "loginCheckJs": "",
            "loginUi": "",
            "loginUrl": "",
            "name": `Azure  ${voice} ${voiceStyle} pitch: ${pitch} rate:${rate}`,
            "url": `${origin}/audio?text={{speakText}}&rate=${rate}&pitch=${pitch}&voice=${voice}&voiceStyle=${voiceStyle},{"method":"GET"}`,
        }

        return new Response(JSON.stringify(dataJson), {
            headers: {
                "content-type": "application/json;charset=UTF-8",
                "Access-Control-Allow-Origin": "*",
                "Access-Control-Allow-Credentials": "true",
                "Access-Control-Allow-Headers": "*",
                "Access-Control-Allow-Methods": "*",
                "ip": `Access cloudflare's ip:${clientIP}`
            },
        })



    } else if (url.pathname == "/sourcereader") {
        const origin = url.origin
        const params = new URLSearchParams(url.search);
        // 獲取查詢參數中的文本
        // const text = params.get("text");
        // 獲取查詢參數中的語速
        const rate = params.get("rate");
        // 獲取查詢參數中的音高
        const pitch = params.get("pitch");
        // 獲取查詢參數中的音色
        const voice = params.get("voice");
        // 獲取查詢參數中的音色風格
        const voiceStyle = params.get("voiceStyle");

        const dataJson = [{
            "customOrder": 100,
            "id": Date.now(),
            "lastUpdateTime": Date.now(),
            "name": ` ${voice} ${voiceStyle} pitch: ${pitch} rate:${rate}`,
            "url": `${origin}/audio?text={{speakText}}&rate=${rate}&pitch=${pitch}&voice=${voice}&voiceStyle=${voiceStyle},{"method":"GET"}`,
        }]
        return new Response(JSON.stringify(dataJson), {
            headers: {
                "content-type": "application/json;charset=UTF-8",
                "Access-Control-Allow-Origin": "*",
                "Access-Control-Allow-Credentials": "true",
                "Access-Control-Allow-Headers": "*",
                "Access-Control-Allow-Methods": "*",
                "ip": `Access cloudflare's ip:${clientIP}`
            },
        })
    }
    else {
        return new Response("page", {
            headers: {
                "content-type": "text/html;charset=UTF-8",
                "Access-Control-Allow-Origin": "*",
                "Access-Control-Allow-Credentials": "true",
                "Access-Control-Allow-Headers": "*",
                "Access-Control-Allow-Methods": "*",
                "ip": `Access cloudflare's ip:${clientIP}`
            },
        })
    }

}

如果想要修改頁面的樣式,可以更改下面這行代碼中的 URL,修改為自己調整後的前端頁面 URL 地址即可:

const html = await fetch("https://files.chiloh.net/azure-tts.html")

創造正在變得越來越簡單,如果你有好的想法,Just do it !!

載入中......
此文章數據所有權由區塊鏈加密技術和智能合約保障僅歸創作者所有。