【音視頻】基於聲網實時音視頻能力的音視頻質量體系建設

{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在音視頻領域帶了多年，發現人們往往會覺得，做Android/iOS業務APP或者SDK開發，不如做底層音視頻編碼解碼/網絡傳輸有挑戰；同理，覺得服務端做業務系統開發，不如做流媒體開發的有挑戰。其實這是一個誤區，對於做工程技術（非算法研究）的研發而言，挑戰的區別根本不是做“上層”還是做“底層”的技術，而是“規模”，一個產品，支撐的業務數量和線上用戶規模上去了，整個產品方方面面都是挑戰：穩定性、容錯性、性能優化，監控告警、降級方案、資源成本、排障和運維效率，測試效率，服務效率等，很多很多點，都是對整條鏈路的每個模塊都提出了更高的要求，更大的挑戰，我們長遠的規劃和技術成長也都是源於如何應對這種規模帶來的挑戰。因此，對於工程師，我們更應該看重和追求的是如何做大產品規模，如何加入有業務規模的研發團隊，而不是片面的覺得“底層開發”更高級，“上層開發”沒挑戰。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這是一位從事音視頻領域多年的大佬的感悟，作爲一個同樣在音視頻行業摸爬滾打多年的打工人，表示深深的贊同。剛參加工作時確實有過音視頻開發”高端“、”有挑戰“、”有技術含量“，隨着工作經驗的積累，以及經手業務規模的提升，對”業務“和”規模“越來越認同。隨着規模的提升，穩定性、性能、監控等越發顯得重要，尤其在音視頻領域，不僅是實時性的挑戰，還有問題定位，業務止損方方面的考慮。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"本文以聲網提供的實時音視頻能力爲例，探討音視頻質量體系的建設。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"1. 音視頻質量評價模型理論","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"質量評估的種類分爲兩種：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"主觀質量評估：利用人眼作爲對象來進行的質量評判和得到MOS分值","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"客觀質量評估：利用數學模型，模擬主觀質量評估的方法，無需人的介入","attrs":{}}]}]}],"attrs":{}},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"1.1 主觀質量評估","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在國際標準中，統一使用MOS值來評價系統接收到的經過壓縮後的音視頻主觀體驗。MOS（Mean Opinion Score）主觀平均意見分。","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/11/11d3c05ca2edd091c7d8e3f8cef508bf.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"主觀體驗的一般流程爲：","attrs":{}}]},{"type":"numberedlist","attrs":{"start":1,"normalizeStart":1},"content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":1,"align":null,"origin":null},"content":[{"type":"text","text":"選取測試視頻","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":2,"align":null,"origin":null},"content":[{"type":"text","text":"衡量音視頻質量的方案","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":3,"align":null,"origin":null},"content":[{"type":"text","text":"設計主觀評估實驗","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":4,"align":null,"origin":null},"content":[{"type":"text","text":"看視頻打分（MOS分）","attrs":{}}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"主觀評估的標準有哪些呢？","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":4},"content":[{"type":"text","text":" 1.1.1 BT.500標準","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"BT.500標準《電視圖像質量的主觀評估方法》主要應用於電視圖像的質量評估，對主觀測試方法進行了以下規定：","attrs":{}}]},{"type":"numberedlist","attrs":{"start":1,"normalizeStart":1},"content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":1,"align":null,"origin":null},"content":[{"type":"text","text":"觀察條件：表一列出了BT.1788建議的觀測條件，觀測距離和水平觀測角度BT.500建議如表二","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":2,"align":null,"origin":null},"content":[{"type":"text","text":"源信號：源信號直接提供基準圖形以及測試中的系統的輸入，源片段的質量應儘可能高。","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":3,"align":null,"origin":null},"content":[{"type":"text","text":"測試材料選擇：測試場景的數目和類型對解釋主觀評估的結果而言是至關重要的。對於特定的測試問題需要選擇相應的測試材料。","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":4,"align":null,"origin":null},"content":[{"type":"text","text":"觀測者：觀測者的數量應達到15人以上，他們應當不是專家，在某種意義上，他們與圖形質量沒有直接利害關係，只是作爲其日常工作的一部分，並且他們不是經驗豐富的評估者。但是對於特殊的測試需求，也可以使用專家觀測者。","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":5,"align":null,"origin":null},"content":[{"type":"text","text":"評估說明：應仔細向評估者介紹評估方法、損傷類型或可能出現的質量因子、等級評定尺度、時間安排等。","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":6,"align":null,"origin":null},"content":[{"type":"text","text":"評估方法","attrs":{}}]}]}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"雙刺激損傷尺度（double-stimulus impairment scale ,DSIS）方法","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"雙刺激連續質量尺度（double-stimulus continuous quality-scale,DSCQS）方法","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"單刺激（Single-stimulus,SS）方法","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"刺激－比較（Stimulus-comparison,SC）方法","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"多媒體視頻質量（Subjective Assessment of Multimedia Video Quality,SAMVIQ）的主觀評估","attrs":{}}]}]}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"表一","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/39/39c795e0ecc9c86925eaadfbdded3a03.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"表二","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/c5/c534d5e59f3b974736b315b34bf69b91.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":4},"content":[{"type":"text","text":"1.1.2 P.910標準","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"P.910標準《多媒體應用的主觀性視頻質量評價方法》主要應用於多媒體應用，例如：視頻會議和相關領域、通信領域，也用於算法的決策、試聽系統的性能等級劃分等。描述了多媒體應用單向整體視頻質量的非交互主觀評估方法。從主觀評測角度該標準描述了4種測試方法：","attrs":{}}]},{"type":"numberedlist","attrs":{"start":1,"normalizeStart":1},"content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":1,"align":null,"origin":null},"content":[{"type":"text","text":"Absolute category rating(ACR) 絕對種類評分；","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":2,"align":null,"origin":null},"content":[{"type":"text","text":"Absolute category rating with hidden reference(ACR-HR) 帶隱藏參考的絕對種類評分；","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":3,"align":null,"origin":null},"content":[{"type":"text","text":"Degradation category rating(DCR) 損傷種類評分；","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":4,"align":null,"origin":null},"content":[{"type":"text","text":"Pair comparison method(PC) 成對比較法。","attrs":{}}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"該標準提到的兩個指標很有參考意義：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"空域感知信息(spatial perceptual information)計算是將視頻中每一幀的亮度平面經過Sobel濾波，濾波器是用來提取並集中高頻成分，將低頻成分過濾掉，再對每一個像素求標準差，進而得到一個時間序列的空域信息，最後這個序列中的最大值用來表達場景內容的空域信息。如果SI值高，表示在一幀內有大量的細節信息，如果該序列是黑屏或者當圖像輪廓較爲模糊，SI指標會是0。","attrs":{}}]}]}],"attrs":{}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/56/56da5a5c669e0ae828c02e8e188eb364.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"感知信息(temporal perceptual information)基於運動差異特徵，即亮度平面相同位置的像素值之差，最後在整個時間域整個空間內標準差最大值用來表達場景內容的時域信息。畫面中運動內容多的TI值更高。如果TI值高，表示場景發生改變頻繁，如果是0表示測試序列是靜止的。","attrs":{}}]}]}],"attrs":{}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/9d/9d1ff2e4148f7e57509d3b6f4ffb6de3.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"雖然該ITU標準是用來主觀打分的，但SI和TI這兩個指標也可以用來做部分參考模型，即比較一下編碼前後SI、TI變化情況，SI變化大，說明編碼器丟失了很多圖像細節紋理信息，TI變化大，說明編碼器丟失了很多運動信息。但這種比較不能經過有損傷的傳輸系統，否則因數據包丟解碼出來有馬賽克的畫面其SI或TI可能比原始圖像還大。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":4},"content":[{"type":"text","text":"1.1.3 GYT 340-2020、 GYT 341-2020 標準","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"GYT 340-2020《超高清晰度電視圖像質量主觀評價方法 雙刺激連續質量標度法》與GYT 341-2020《基於IP傳輸的地面數字電視廣播單頻網組網技術規範》是國家廣播電視總局批准的中華人民共和國廣播電視和網絡視聽推薦性行業標準。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":4},"content":[{"type":"text","text":"1.1.4 總結","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"主觀質量標準對環境、設備、素材、人員、流程、結果有要求。由於是人眼測試，主觀評價優點是準確性高；缺點是長時間主觀評估，人易疲勞，人的主觀性使得測試結果可重複性差，多人長時間的參與開銷大。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"主觀評估的成本高，無法大批量進行評估，只能用來少量評估場景、或者客觀評估的的輔助。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"MOS分的最主要價值是指導建設客觀評估體系。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"1.2 客觀質量評估","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"客觀質量評估是利用數學模型來模擬主觀評估結果，同樣的輸入參數不變，每次評估結果一樣。客觀評估分爲以下三類：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"FR（Full Reference）全參考評估","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"RR（Reduced Reference）部分參考評估","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"NR（No Reference）：無參考評估","attrs":{}}]}]}],"attrs":{}},{"type":"heading","attrs":{"align":null,"level":4},"content":[{"type":"text","text":"1.2.1 全參考評估","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"原始圖像和壓縮後 圖像在同一位置的每個像素的比較結果。","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"傳統點運算（對於圖像的位運算比較）：","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"MSE-均方誤差（MSE，mean squared error）：首先計算原始圖像和失真像象素差值的均方值，然後通過均方值的大小來確定失真圖像的失真程度","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"PSNRPSNR-峯值信噪比（PSNR，peak signal to noise rate）：衡量圖像質量的重要指標，基於通信理論而提出，是最大信號量與噪聲強度的比值。","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"自然視覺特徵（模型中添加人眼視覺HSV比較敏感的元素的分析，如亮度）：","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"SSIM 基於圖像結構相似度的質量評價方法。有爭議，與主觀評估不太一致。","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"VQM 利用統計學原理模擬實際的人眼視覺系統，提取人眼能感知的圖像特徵值的基礎上，計算出視頻質量的客觀評價。","attrs":{}}]}]}],"attrs":{}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/da/da65b78e959ead26d09f4bec66b88bd3.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"優點：最準確","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"缺點（無法在遠程實時監測視頻質量）：","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"實現成本高","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"原始視頻的傳輸成本高","attrs":{}}]}]}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"主要應用於視頻編解碼系統的性能測試，如H.264與H.265的編碼對比測試。","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":4},"content":[{"type":"text","text":"1.2.2 部分參考評估","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"原始視頻和解碼視頻中提取一些特徵值進行比較和評估結果。","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/ce/ce381634f23d1d165686b7f65a86da71.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"部分參考評估需要原始視頻，準確度降低","attrs":{}}]},{"type":"heading","attrs":{"align":null,"level":4},"content":[{"type":"text","text":"1.2.3 無參考評估","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/a6/a6d423f1c7cd11601887b99f9b63ec9a.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"缺點：不準確","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"優點（特徵提取+機器學習）：","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"不需要原始視頻","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":1,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"實現成本低","attrs":{}}]}]}],"attrs":{}},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"1.3 主、客觀評估的結合","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/f2/f2e58f9823423436629dfd5effbc90a6.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"主觀評價：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"準確性高","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"實施成本高","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":" 可重複性差","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"無法大批量評估","attrs":{}}]}]}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"客觀質量評估：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"準確性依賴於評估的數學模型","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"每次評估結果一樣","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"可大批量評估","attrs":{}}]}]}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"結合兩種質量評估：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"主觀評估的準確性高的優點，來檢驗客觀評估的數據模型","attrs":{}}]}]}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"有點類似於機器學習中的標註和訓練來提高預測準確度。客觀評估的兩個重點：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"指標","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"數學模型","attrs":{}}]}]}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"2. 音視頻質量評價實踐","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在實時音視頻中我們圍繞着畫質、聲音、流暢度/卡頓建立評價體系。根據核心指標拆分出不同的維度，如：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"拉流成功率","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"百秒卡頓時長","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"百秒卡頓次數","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"端到端延遲","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"首幀時長","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":" ...","attrs":{}}]}]}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"在實際的生產環境我們很難拿到原始視頻，所以只能基於NR的客觀評估模型進行評估，然後結合MOS方法主觀抽樣評價打造能反映用戶真實體驗感受的指標模型。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"接下來我們就需要創建我們的數學模型和進行數據採集。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"2.1 客觀評價體系數學模型","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"數學模型模擬人眼觀看視頻的主觀體驗，輸入視頻的指標數據，輸出客觀評分的結果。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"我們這裏採用","attrs":{}},{"type":"text","marks":[{"type":"color","attrs":{"color":"#F5222D","name":"red"}},{"type":"strong","attrs":{}}],"text":"差體驗扣分","attrs":{}},{"type":"text","text":"的算法。扣分算法是指在每次遇到的差體驗，按一定的規則進行扣分。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"總分公式：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"總分=100分-（a1+a2+a3...+am)","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"a1,a2..,am是每次連續的差體驗扣分。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"α1 = 100 * C * ∑( Max(β1,β2,...,βj) ) * / ∑(0,n)","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"C:頻次係數，同時長的差體驗，短時間，多次的差體驗比一次長時間差體驗感覺越差","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"β：指標係數，根據數值變化，同一時間，發生多個差體驗（比如：卡頓、幀率低），按照更差的指標係數扣分","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"∑ (Max(β1,β2,...,βj))：連續差體驗時，最高的指標係數值的相加","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"n：播放總時長","attrs":{}}]}]}],"attrs":{}},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"2.2 指標收集","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"有了算法模型，我們該去關心哪些指標，怎麼收集指標呢？在聲網的控制檯水晶球中我們可以看到聲網官方提供的指標信息：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"概覽","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/a3/a318f78334e1a86ceea416dc69dabba7.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"音頻","attrs":{}}]},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/d4/d44099dba7e77c7a5224cf2327e571ef.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/16/16df6c577f9fbde548a7b48166513bff.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong","attrs":{}}],"text":"視頻","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/38/38ff72274bcaf167b11cd4fd4dbdf268.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/a3/a3a6a7ede92357a391d1384ca9c53a3f.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/73/7309642c1e1b49b822585606407cb913.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"我們看到的指標有：","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"設備狀態","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"音頻採集音量","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"音頻上行和網絡丟包","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"音頻解碼卡頓","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"視頻發送/接收分辨率","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"視頻發送/接收幀率","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"視頻上行和網絡丟包","attrs":{}}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"視頻下行和端到端丟包","attrs":{}}]}]}],"attrs":{}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"我們在使用聲網服務時端上如何採集這些信息？以聲網視頻通話SDK爲例，agora SDK提供了數據統計事件，加入頻道後，SDK每隔2秒自動觸發回調：","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/ca/ca0e537cbe9d9cfd770f0694ab295f2a.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"回調中包含了視頻寬高、接收碼率、遠端視頻解碼器的輸出幀率、遠端視頻渲染器的輸出幀率、遠端視頻在網絡對抗後的丟包率、遠端用戶在加入頻道後發生視頻卡頓的累計時長、遠端用戶在加入頻道後發生視頻卡頓的累計時長佔視頻總有效時長的百分比 、視頻總有效時長、遠端視頻流的累計發佈時長、實際發送碼率、實際發送幀率、本地編碼器的輸出幀率、本地渲染器的輸出幀率、當前編碼器的目標編碼碼率、當前編碼器的目標編碼幀率、自上次統計後本地視頻質量的自適應情況、視頻編碼碼率、視頻發送的幀數，累計值、弱網對抗前本地客戶端到 Agora 邊緣服務器的視頻丟包率 (%)、本地視頻採集幀率、本地採集的畫質亮度級別、遠端用戶發送的音頻流質量、音頻發送端到接收端的網絡延遲、接收端到網絡抖動緩衝的網絡延遲、統計週期內的遠端音頻流的丟幀率、統計週期內接收到的遠端音頻採樣率、接收流在統計週期內的平均碼率、遠端用戶在加入頻道後發生音頻卡頓的累計時長、遠端用戶在加入頻道後發生音頻卡頓的累計時長佔音頻總有效時長的百分比、遠端用戶在音頻通話開始到本次回調之間的有效時長、遠端音頻流的累計發佈時長、接收遠端音頻時，本地用戶的主觀體驗質量、接收遠端音頻時，本地用戶主觀體驗質量較差的原因、統計週期內，Agora 實時音頻 MOS（平均主觀意見分）評估方法對接收到的遠端音頻流的質量評分。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":3},"content":[{"type":"text","text":"2.3 評分模型實現","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https://static001.geekbang.org/infoq/4a/4ae84ace69fc46fde2329fd4dfff5852.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"sdk回調收集上傳指標信息，藉助大數據等數據處理平臺做數據的實時計算，從而完成基於聲網實時音視頻能力的質量體系建設，並形成監控報警工具。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"3.總結","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"本文介紹了音視頻質量體系模型理論：主觀質量評估與客觀質量評估，並基於聲網視頻通話SDK提供的統計能力，構建基於扣分模型的質量體系評價模型和實現方案。","attrs":{}}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"4.參考","attrs":{}}]},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"《電視圖像質量的主觀評估方法》：","attrs":{}},{"type":"link","attrs":{"href":"https://www.itu.int/rec/R-REC-BT.500","title":"","type":null},"content":[{"type":"text","text":"https://www.itu.int/rec/R-REC-BT.500","attrs":{}}]}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"《多媒體應用的主觀性視頻質量評價方法》：","attrs":{}},{"type":"link","attrs":{"href":"https://www.itu.int/rec/T-REC-P.910","title":"","type":null},"content":[{"type":"text","text":"https://www.itu.int/rec/T-REC-P.910","attrs":{}}]}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"GYT 340-2020《超高清晰度電視圖像質量主觀評價方法 雙刺激連續質量標度法》","attrs":{}},{"type":"link","attrs":{"href":"http://www.nrta.gov.cn/module/download/downfile.jsp?spm=chekydwncf.0.0.1.YworBf&classid=0&showname=GYT%20340-2020%E3%80%8A%E8%B6%85%E9%AB%98%E6%B8%85%E6%99%B0%E5%BA%A6%E7%94%B5%E8%A7%86%E5%9B%BE%E5%83%8F%E8%B4%A8%E9%87%8F%E4%B8%BB%E8%A7%82%E8%AF%84%E4%BB%B7%E6%96%B9%E6%B3%95%20%E5%8F%8C%E5%88%BA%E6%BF%80%E8%BF%9E%E7%BB%AD%E8%B4%A8%E9%87%8F%E6%A0%87%E5%BA%A6%E6%B3%95%E3%80%8B.pdf&filename=6f988667cc924c32b05fefa71eb0ca17.pdf","title":"","type":null},"content":[{"type":"text","text":"http://www.nrta.gov.cn/module/download/downfile.jsp?spm=chekydwncf.0.0.1.YworBf&classid=0&showname=GYT%20340-2020%E3%80%8A%E8%B6%85%E9%AB%98%E6%B8%85%E6%99%B0%E5%BA%A6%E7%94%B5%E8%A7%86%E5%9B%BE%E5%83%8F%E8%B4%A8%E9%87%8F%E4%B8%BB%E8%A7%82%E8%AF%84%E4%BB%B7%E6%96%B9%E6%B3%95%20%E5%8F%8C%E5%88%BA%E6%BF%80%E8%BF%9E%E7%BB%AD%E8%B4%A8%E9%87%8F%E6%A0%87%E5%BA%A6%E6%B3%95%E3%80%8B.pdf&filename=6f988667cc924c32b05fefa71eb0ca17.pdf","attrs":{}}]}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"GYT 341-2020《基於IP傳輸的地面數字電視廣播單頻網組網技術規範》","attrs":{}},{"type":"link","attrs":{"href":"http://www.nrta.gov.cn/module/download/downfile.jsp?spm=chekydwncf.0.0.1.YworBf&classid=0&showname=GYT%20341-2020《基於IP傳輸的地面數字電視廣播單頻網組網技術規範》.pdf&filename=9a308c5153304096a9808da0131c11c9.pdf","title":"","type":null},"content":[{"type":"text","text":"http://www.nrta.gov.cn/module/download/downfile.jsp?spm=chekydwncf.0.0.1.YworBf&classid=0&showname=GYT%20341-2020《基於IP傳輸的地面數字電視廣播單頻網組網技術規範》.pdf&filename=9a308c5153304096a9808da0131c11c9.pdf","attrs":{}}]}]}]}],"attrs":{}}]}