Use Case Intelligence RSS Search use cases Search
Baidu AI Cloud Media and Entertainment China Translated from Chinese

CCTV.com

Enterprise Knowledge Search

CCTV.com needed a better way to surface the value of its huge video archive because title- and keyword-based retrieval could not expose enough scene-level meaning, leaving many useful videos hard to find.

Read the original story Explore all use cases
Resource retrieval efficiency improved 90%.

Reported impact

Business outcome

CCTV.com made its video catalog easier to search, improved retrieval efficiency, and gave users a more natural voice- and semantics-driven way to discover media assets.

  • Customer Experience
  • Productivity Improvement

Solution

CCTV.com and Baidu AI Cloud built the 'AI Helps You Find' intelligent search system using speech recognition and multimodal media-retrieval capabilities. Instead of depending only on titles, descriptions, and manually assigned keywords, the system analyzes video content with face recognition, OCR, natural-language processing, and knowledge-graph techniques to extract many dimensions of tags from the media itself. Those tags then drive more expressive search and recommendation behaviors, letting users query from multiple angles and still find relevant content. Voice input is layered on top so users can issue commands without typing and receive results quickly. The result is an AI media-search architecture in which speech input, multimodal understanding, and tag-driven retrieval work together to unlock previously 'cold' assets in CCTV.com's archive and turn a large media catalog into a more accessible interactive knowledge surface for audiences.

Data flow

Users issue spoken or typed requests into the search system, which maps them against AI-generated video tags derived from faces, text, language, and knowledge-graph relationships. Matching results are then returned as search and recommendation outputs that help users locate relevant clips quickly.

Solution architecture

2 components · 2 layers
  1. Ingest
    • Real-Time Speech Recognition Provides voice-command input for the intelligent media-search experience.
  2. Serving
    • Multimodal Media Retrieval Indexes video resources with AI-generated tags and retrieves results across multiple semantic dimensions.

Architecture clues

  • The goal was to unlock value from otherwise hard-to-find archived videos
  • The search system fuses face recognition, OCR, NLP, and knowledge graphs to tag videos across many dimensions
  • Voice commands were added as a first-class interaction layer

Evidence from the source

CCTV.com and Baidu AI Cloud built the 'AI Helps You Find' intelligent search system, and resource retrieval efficiency improved 90% (original: 央视网联合百度智能云打造“AI帮你找”智能搜索系统…资源检索效率提升90%).
The system combines face recognition, text recognition, natural-language processing, and knowledge graphs to extract tags from multiple dimensions of the video (original: 将人脸识别、文字识别、自然语言处理、知识图谱等人工智能技术进行综合…形成不同类型的视频标签).
Traditional video search based on titles, descriptions, and keywords cannot expose all value points in the video, leaving many assets effectively 'frozen' (original: 传统的视频搜索中，用户主要是通过视频标题、简介、关键词等内容，对视频进行搜索…很多视频因此处于被『冷藏』的状态).
Voice input gives users a more convenient command experience and makes it possible to request programs by speaking rather than typing (original: 『语音输入』为用户提供了更便捷的指令发送方式，『动嘴不动手』的节目点播体验).
Transparency Evidence & methodology 90/100 reference value · 89% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value90/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact15/25
    1 metric describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence89%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    4 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. Baidu AI Cloud

    CCTV.com customer story

    First-party
    https://cloud.baidu.com/customer/case/yangshi.html
    Source type
    Vendor-published customer story
    Published
    May 15, 2025
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

  • Published on source page May 15, 2025 · Baidu AI Cloud · First-party
    Open date source
Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Related

Similar use cases

See all like this
Baidu AI Cloud May 15, 2025 The Paper Media and Entertainment China The Paper wanted to use generative AI in news products, but needed a way to answer readers' questions without hallucinations while preserving the accuracy standards of professional journalism. The assistant is grounded on more than 27 million articles and nearly 35 billion characters of original reporting. Enterprise Knowledge Search AI Q&A Assistant · Qianfan AppBuilder · Qianfan ModelBuilder Baidu AI Cloud May 15, 2025 Tianmai Juyuan Media and Entertainment China Tianmai Juyuan needed to organize and search a massive broadcast-video archive quickly enough to support timely publishing, short-video creation and intelligent content retrieval. MCA analysis efficiency improved 3-5x over manual tagging. Enterprise Knowledge Search BOS Object Storage · Media Content Analysis (MCA) · Micro-Editing Platform Lenovo Oct 13, 2026 Formula E Media and Entertainment global Formula E wanted fan experiences that feel immediate, immersive, personalized, and participatory, but its broadcast and event experience was limited by inconsistent onboard video quality, information spread across… Developed in under 30 days. Enterprise Knowledge SearchReal-time Analytics AI Fan Experience Guide · AI Video Optimization · AR broadcast graphics