テキスト→音声変換の「Text To Speech API」の使い方と、2017年3月現在のWatsonとの違い：認識系API活用入門（3）（2/3 ページ）

// ----------------------------------------------
// Tokenを取得する処理
// ----------------------------------------------
string GetAccessToken()
{
    try
    {
        //取得したBing Speech API の Key
        string subscriptionKey = "xxxxxxxxxxxxxxxxxxxxxxxx";
 
        WebRequest request = WebRequest.Create("https://api.cognitive.microsoft.com/sts/v1.0/issueToken");
 
        //リクエスト設定
        request.Method = "POST"; // メソッドはPOST
        request.Headers.Add("Ocp-Apim-Subscription-Key", subscriptionKey);
        request.ContentLength = 0;
 
        // リクエストを投げる
        WebResponse response = request.GetResponse();
        Stream responseStream = response.GetResponseStream();
 
        //　レスポンスを受け取る
        StreamReader streamReader = new StreamReader(responseStream);
 
        //　Tokenを取得した時間を取得
        tokenRefreshedTime = DateTime.Now;
 
        // レスポンス（Token）を返す
        return streamReader.ReadToEnd();
    }
    catch (WebException webException)
    {
        Stream responseStream = webException.Response.GetResponseStream();
        StreamReader streamReader = new StreamReader(responseStream);
        tblkMessage.Text = streamReader.ReadToEnd();
        return null;
    }
    catch (Exception exception)
    {
        tblkMessage.Text = exception.Message;
        return null;
    }
}

<speak version='1.0' xml:lang='en-US'><voice xml:lang='en-US' xml:gender='Female' name='Microsoft Server Speech Text to Speech Voice (en-US, ZiraRUS)'>Microsoft Bing Voice Output API</voice></speak>

// ----------------------------------------------
// 音声ファイルを取得する処理
// ----------------------------------------------
string GetTextToSpeechAudio(string pAccessToken, string pLanguage, string pGender, string pText, string pAudioFormat)
{
    try
    {
        string url = "https://speech.platform.bing.com/synthesize";
        string speakerName = "";
        switch (pLanguage)
        {
        case "en-US":
            if (pGender == "Female") speakerName = "ZiraRUS"; else speakerName = "BenjaminRUS";
            break;
        case "ja-JP":
            if (pGender == "Female") speakerName = "Ayumi, Apollo"; else speakerName = "Ichiro, Apollo";
                break;
        }
 
        // インプットパラメーター（SSML）作成
        string requestSSML =
            string.Format(
                "<speak version='1.0' xml:lang='{0}'><voice xml:lang='{0}' xml:gender='{1}' name='Microsoft Server Speech Text to Speech Voice ({0}, {2})'>{3}</voice></speak>",
                pLanguage,
                pGender,
                speakerName,
                pText
            );
 
        // リクエスト作成
        HttpWebRequest request = WebRequest.CreateHttp(url);        
        request.Method = "POST";
        request.ContentType = "application/ssml+xml";
        if (pAudioFormat == "WAV")
        {
            request.Headers.Add("X-Microsoft-OutputFormat", "riff-16khz-16bit-mono-pcm");
        }
        else
        { 
            request.Headers.Add("X-Microsoft-OutputFormat", "audio-16khz-128kbitrate-mono-mp3");
        }
        request.Headers.Add("X-Search-AppId", (Guid.NewGuid().ToString()).Replace("-", ""));
        request.Headers.Add("X-Search-ClientID", (Guid.NewGuid().ToString()).Replace("-", ""));
        request.Headers.Add("Authorization", "Bearer " + pAccessToken);
        request.UserAgent = "SpeechToTextDemoApp";
 
        // パラメーターをUTF8でエンコード
        byte[] bytes = System.Text.Encoding.UTF8.GetBytes(requestSSML);
        request.ContentLength = bytes.Length;
 
        // パラメーターをリクエストに追加
        System.IO.Stream s = request.GetRequestStream();
        s.Write(bytes, 0, bytes.Length);
        s.Close();
 
        // リクエスト送信／レスポンス取得
        WebResponse webResponse = request.GetResponse();
 
        // レスポンスをストリームに変換
        Stream responseStream = webResponse.GetResponseStream();
        if (responseStream == null)
        {
            throw new Exception("responseStreamがNULLです。");
        }
 
        // レスポンスの中身を音声ファイルとして保存
        string audioFile = "";
        if (pAudioFormat == "WAV")
        {
            audioFile = System.Environment.GetFolderPath(Environment.SpecialFolder.Personal) + @"\TextToSpeech.wav";
        }
        else
        {
            audioFile = System.Environment.GetFolderPath(Environment.SpecialFolder.Personal) + @"\TextToSpeech.mp3";
        }
        FileStream fs = new FileStream(audioFile, FileMode.Create, FileAccess.Write);
 
        byte[] readData = new byte[1024];
        while(true)
        {
            // データを読み込む
            int readSize = responseStream.Read(readData, 0, readData.Length);
            if (readSize == 0)
            {
                // 全てのデータを読み込んだら抜ける
                break;
            }
            // 書き込む
            fs.Write(readData, 0, readSize);
        }
 
        fs.Close();
        responseStream.Close();
 
        return audioFile;
 
    }
    catch (WebException webException)
    {
        Stream responseStream = webException.Response.GetResponseStream();
                StreamReader streamReader = new StreamReader(responseStream);
        tblkMessage.Text = streamReader.ReadToEnd();
        return null;
    }
    catch (Exception exception)
    {
        tblkMessage.Text = exception.Message;
        return null;
    }
}