Quickstart: Use Azure Content Understanding na Foundry Tools REST API

Este guia de início rápido mostra como usar a API REST de Compreensão de Conteúdo para obter dados estruturados de conteúdo multimodal em arquivos de documento, imagem, áudio e vídeo.

Pré-requisitos

Para começar, você precisa de uma assinatura ativa do Azure. Se você não tiver uma conta do Azure, crie uma gratuitamente.
Depois de ter a sua subscrição Azure, crie um recurso Microsoft Foundry no portal Azure. Certifique-se de criá-lo em uma região suportada.
Configure implementações de modelos padrão para o seu recurso de Compreensão de Conteúdo. Definir os padrões cria uma ligação aos modelos Foundry que usa para pedidos de Compreensão de Conteúdo. Escolha um dos métodos seguintes:
- Portal
- API REST
1. Vá à página de definições de Compreensão de Conteúdo
2. Selecione o botão "+ Adicionar recurso" no canto superior esquerdo
3. Selecione o recurso Foundry que pretende usar e clique em Seguinte, depois Guardar
  - Certifique-se de deixar "Ativar a implantação automática para modelos obrigatórios caso não existam predefinições disponíveis." assinalado. Isto assegura que o seu recurso está configurado integralmente com os modelos requisitados GPT-4.1, GPT-4.1-mini e text-embedding-3-large. Diferentes analisadores pré-montados requerem modelos diferentes.
Ao seguir estes passos, estabelece uma ligação entre Compreensão de Conteúdo e modelos Foundry no seu recurso Foundry.
1. Crie implementações do Modelo Foundry do GPT-4.1, GPT-4.1-mini e text-embedding-3-large no seu recurso Foundry. Para detalhes sobre como implementar estes modelos, consulte Criar implementações de modelos no portal Microsoft Foundry. Diferentes analisadores pré-montados requerem modelos diferentes, por isso tens de implementar os três.
2. Defina as implementações padrão do modelo ao nível dos recursos.
  
  Antes de executar o seguinte comando cURL, faça as seguintes alterações na solicitação HTTP:
  - Substitua {endpoint} e {key} pelos valores correspondentes da sua instância Foundry no portal Azure.
  - Substitua {myGPT41Deployment}, {myGPT41MiniDeployment}, e {myEmbeddingDeployment} pelos nomes reais de implementação dos seus modelos a partir do seu recurso Foundry.
```
curl -i -X PATCH "{endpoint}/contentunderstanding/defaults?api-version=2025-11-01" \
  -H "Ocp-Apim-Subscription-Key: {key}" \
  -H "Content-Type: application/json" \
  -d '{
        "modelDeployments": {
          "gpt-4.1": "{myGPT41Deployment}",
          "gpt-4.1-mini": "{myGPT41MiniDeployment}",
          "text-embedding-3-large": "{myEmbeddingDeployment}"
        }
      }'
```
Ao seguir estes passos, estabelece uma ligação entre Compreensão de Conteúdo e modelos Foundry no seu recurso Foundry.
Neste guia, usamos a ferramenta de linha de comando cURL. Se ele não estiver instalado, você poderá baixar a versão apropriada para seu ambiente de desenvolvimento.

Introdução a um analisador pré-construído

Os analisadores definem como seu conteúdo é processado e os insights que são extraídos. Oferecemos analisadores pré-construídos para casos de uso comuns. Você pode personalizar analisadores pré-construídos para melhor atender às suas necessidades específicas e casos de uso. Este início rápido utiliza analisadores pré-definidos de faturas, imagens, áudio e vídeo para ajudá-lo a começar.

Envie um ficheiro para análise

Antes de executar o seguinte comando cURL, faça as seguintes alterações na solicitação HTTP:

Substitua {endpoint} e {key} pelos valores correspondentes da sua instância Foundry no portal Azure.

Pedido POST

Este exemplo utiliza o prebuilt-invoice analisador para extrair dados estruturados de um documento de fatura.

curl -i -X POST "{endpoint}/contentunderstanding/analyzers/prebuilt-invoice:analyze?api-version=2025-11-01" \
  -H "Ocp-Apim-Subscription-Key: {key}" \
  -H "Content-Type: application/json" \
  -d '{
        "inputs":[{"url": "https://github.com/Azure-Samples/azure-ai-content-understanding-python/raw/refs/heads/main/data/invoice.pdf"}]
      }'

Este exemplo utiliza o prebuilt-imageSearch analisador para gerar uma descrição da imagem.

curl -i -X POST "{endpoint}/contentunderstanding/analyzers/prebuilt-imageSearch:analyze?api-version=2025-11-01" \
  -H "Ocp-Apim-Subscription-Key: {key}" \
  -H "Content-Type: application/json" \
  -d '{
        "inputs":[
          {
            "url": "https://github.com/Azure-Samples/azure-ai-content-understanding-python/raw/refs/heads/main/data/pieChart.jpg"
          }          
        ]
      }'

Este exemplo utiliza o prebuilt-audioSearch analisador para extrair a transcrição áudio, gerar um resumo e realizar a rotulagem dos altifalantes.

curl -i -X POST "{endpoint}/contentunderstanding/analyzers/prebuilt-audioSearch:analyze?api-version=2025-11-01" \
  -H "Ocp-Apim-Subscription-Key: {key}" \
  -H "Content-Type: application/json" \
  -d '{
        "inputs":[
          {
            "url": "https://github.com/Azure-Samples/azure-ai-content-understanding-python/raw/refs/heads/main/data/audio.wav"
          }          
        ]
      }'

Este exemplo utiliza o prebuilt-videoSearch analisador para extrair fotogramas-chave, transcrições e segmentos de capítulos de vídeo.

curl -i -X POST "{endpoint}/contentunderstanding/analyzers/prebuilt-videoSearch:analyze?api-version=2025-11-01" \
  -H "Ocp-Apim-Subscription-Key: {key}" \
  -H "Content-Type: application/json" \
  -d '{
        "inputs":[
          {
            "url": "https://github.com/Azure-Samples/azure-ai-content-understanding-python/raw/refs/heads/main/data/FlightSimulator.mp4"
          }          
        ]
      }'

Resposta POST

O cabeçalho de resposta inclui um Operation-Location campo, que utiliza para recuperar os resultados da operação de análise assíncrona.

HTTP/1.1 202 Accepted
Transfer-Encoding: chunked
Content-Type: application/json
request-id: aaa-bbb-ccc-ddd
x-ms-request-id: aaa-bbb-ccc-ddd
Operation-Location: {endpoint}/contentunderstanding/analyzerResults/{request-id}?api-version=2025-11-01
api-supported-versions: 2024-12-01-preview,2025-05-01-preview,2025-11-01
x-envoy-upstream-service-time: 800
apim-request-id: {request-id}
Strict-Transport-Security: max-age=31536000; includeSubDomains; preload
x-content-type-options: nosniff
x-ms-region: West US
Date: Fri, 31 Oct 2025 05:30:17 GMT
Connection: close

Importante

Copie o URL de Localização da Operação a partir do cabeçalho da resposta. Vai usar este URL na próxima etapa para obter os resultados da análise.

Obter resultado da análise

Use o Operation-Location da POST resposta e recupere o resultado da análise.

Pedido GET

curl -i -X GET "{endpoint}/contentunderstanding/analyzerResults/{request-id}?api-version=2025-11-01" \
  -H "Ocp-Apim-Subscription-Key: {key}"

OBTER resposta

A resposta JSON 200 (OK) inclui um status campo que indica o status da operação. Se a operação não estiver concluída, o valor de status é Running ou NotStarted. Nesses casos, você deve enviar a GET solicitação novamente, manualmente ou por meio de um script. Aguarde um intervalo de um segundo ou mais entre as chamadas.

{
  "id": "ce05fb5a-579e-4f0b-afb5-3532bcddeaee",
  "status": "Succeeded",
  "result": {
    "analyzerId": "prebuilt-invoice",
    "apiVersion": "2025-11-01",
    "createdAt": "2025-11-13T20:04:55Z",
    "warnings": [],
    "contents": [
      {
        "path": "input1",
        "markdown": "CONTOSO LTD.\n\n\n# INVOICE\n\nContoso Headquarters\n123 456th St\nNew York, NY, 10001\n\nINVOICE: INV-100\n\nINVOICE DATE: 11/15/2019\n\nDUE DATE: 12/15/2019\n\nCUSTOMER NAME: MICROSOFT CORPORATION...",
        "fields": {
          "AmountDue": {
            "type": "object",
            "valueObject": {
              "Amount": {
                "type": "number",
                "valueNumber": 610,
                "spans": [
                  {
                    "offset": 1522,
                    "length": 7
                  }
                ],
                "confidence": 0.773,
                "source": "D(1,7.3628,8.0459,7.9272,8.0459,7.9272,8.2070,7.3628,8.2070)"
              },
              "CurrencyCode": {
                "type": "string",
                "valueString": "USD"
              }
            }
          },
          "BalanceForward": {
            "type": "object",
            "valueObject": {
              "Amount": {
                "type": "number",
                "valueNumber": 500,
                "spans": [
                  {
                    "offset": 1474,
                    "length": 7
                  }
                ],
                "confidence": 0.901,
                "source": "D(1,7.3628,7.7445,7.9278,7.7467,7.9272,7.9092,7.3622,7.9070)"
              },
              "CurrencyCode": {
                "type": "string",
                "valueString": "USD"
              }
            }
          },
          "BillingAddress": {
            "type": "string",
            "valueString": "123 Bill St, Redmond WA, 98052",
            "spans": [
              {
                "offset": 325,
                "length": 12
              },
              "..."
            ],
            "confidence": 0.712,
            "source": "D(1,0.5805,3.9471,1.2858,3.9478,1.2856,4.1115,0.5803,4.1108);..."
          },
          "BillingAddressRecipient": {
            "type": "string",
            "valueString": "Microsoft Finance",
            "spans": [
              {
                "offset": 307,
                "length": 17
              }
            ],
            "confidence": 0.815,
            "source": "D(1,0.5734,3.7392,1.8060,3.7521,1.8043,3.9201,0.5717,3.9072)"
          },
          "CountryRegion": {
            "type": "string",
            "valueString": "USA"
          },
          "CustomerAddress": {
            "type": "string",
            "valueString": "123 Other St, Redmond WA, 98052",
            "spans": [
              "..."
            ],
            "confidence": 0.744,
            "source": "..."
          },
          "CustomerAddressRecipient": {
            "type": "string",
            "valueString": "Microsoft Corp",
            "spans": [
              "..."
            ],
            "confidence": 0.437,
            "source": "..."
          },
          "CustomerId": {
            "type": "string",
            "valueString": "CID-12345",
            "spans": [
              "..."
            ],
            "confidence": 0.936,
            "source": "..."
          },
          "CustomerName": {
            "type": "string",
            "valueString": "MICROSOFT CORPORATION",
            "spans": [
              "..."
            ],
            "confidence": 0.46,
            "source": "..."
          },
          "CustomerTaxId": {
            "type": "string",
            "confidence": 0.912
          },
          "DueDate": {
            "type": "date",
            "valueDate": "2019-12-15",
            "spans": [
              "..."
            ],
            "confidence": 0.97,
            "source": "..."
          },
          "InvoiceDate": {
            "type": "date",
            "valueDate": "2019-11-15",
            "spans": [
              "..."
            ],
            "confidence": 0.939,
            "source": "..."
          },
          "InvoiceId": {
            "type": "string",
            "valueString": "INV-100",
            "spans": [
              "..."
            ],
            "confidence": 0.733,
            "source": "..."
          },
          "LineItems": {
            "type": "array",
            "valueArray": [
              {
                "type": "object",
                "valueObject": {
                  "Date": {
                    "type": "date",
                    "valueDate": "2021-03-04",
                    "spans": [
                      "..."
                    ],
                    "confidence": 0.894,
                    "source": "..."
                  },
                  "Description": {
                    "type": "string",
                    "valueString": "Consulting Services",
                    "spans": [
                      "..."
                    ],
                    "confidence": 0.589,
                    "source": "..."
                  },
                  "ProductCode": {
                    "type": "string",
                    "valueString": "A123",
                    "spans": [
                      "..."
                    ],
                    "confidence": 0.879,
                    "source": "..."
                  },
                  "Quantity": {
                    "type": "number",
                    "valueNumber": 2,
                    "spans": [
                      "..."
                    ],
                    "confidence": 0.939,
                    "source": "..."
                  },
                  "QuantityUnit": {
                    "type": "string",
                    "valueString": "hours",
                    "spans": [
                      "..."
                    ],
                    "confidence": 0.85,
                    "source": "..."
                  },
                  "TaxAmount": {
                    "type": "object",
                    "valueObject": {
                      "Amount": {
                        "type": "number",
                        "valueNumber": 6,
                        "spans": [
                          "..."
                        ],
                        "confidence": 0.522,
                        "source": "..."
                      },
                      "CurrencyCode": {
                        "type": "string",
                        "valueString": "USD"
                      }
                    }
                  },
                  "TaxRate": {
                    "type": "number",
                    "confidence": 0.915
                  },
                  "TotalAmount": {
                    "type": "object",
                    "valueObject": {
                      "Amount": {
                        "type": "number",
                        "valueNumber": 60,
                        "spans": [
                          "..."
                        ],
                        "confidence": 0.972,
                        "source": "..."
                      },
                      "CurrencyCode": {
                        "type": "string",
                        "valueString": "USD"
                      }
                    }
                  },
                  "UnitPrice": {
                    "type": "object",
                    "valueObject": {
                      "Amount": {
                        "type": "number",
                        "valueNumber": 30,
                        "spans": [
                          "..."
                        ],
                        "confidence": 0.97,
                        "source": "..."
                      },
                      "CurrencyCode": {
                        "type": "string",
                        "valueString": "USD"
                      }
                    }
                  }
                }
              },
              "... (2 additional line items)"
            ]
          }
          /*additional fields omitted*/
        },
        "kind": "document",
        "startPageNumber": 1,
        "endPageNumber": 1,
        "unit": "inch",
        "pages": [
          {
            "pageNumber": 1,
            "angle": 0,
            "width": 8.5,
            "height": 11,
            "words": [
              "... (words omitted for brevity)"
            ],
            "selectionMarks": [],
            "lines": [
              "... (lines omitted for brevity)"
            ],
            "barcodes": [],
            "formulas": []
          }
        ],
        "tables": [
          "... (tables omitted for brevity)"
        ],
        "keyValuePairs": [
          "... (key-value pairs omitted for brevity)"
        ],
        "analyzerId": "prebuilt-invoice",
        "mimeType": "application/pdf"
      }
    ]
  },
  "usage": {
    "documentStandardPages": 1,
    "contextualizationTokens": 2345,
    "tokens": {
      "gpt-4.1-mini-input": 1234,
      "gpt-4.1-mini-output": 567
    }
  }
}

{
  "id": "fe6bb69d-1d6b-4698-a50c-ce798bacdd95",
  "status": "Succeeded",
  "result": {
    "analyzerId": "prebuilt-imageSearch",
    "apiVersion": "2025-11-01",
    "createdAt": "2025-11-13T20:11:50Z",
    "warnings": [],
    "contents": [
      {
        "path": "input1",
        "markdown": "![image](pages/1)\n",
        "fields": {
          "Summary": {
            "type": "string",
            "valueString": "The image is a 3D pie chart illustrating the distribution of hours spent in four different ranges. The largest segment, colored purple, represents 60+ hours at 37.8%. The second largest segment, in teal, represents 50-60 hours at 36.6%. The red segment shows 40-50 hours at 18.9%, and the smallest orange segment represents 1-39 hours at 6.7%. Each segment is labeled with its corresponding range and percentage."
          }
        },
        "kind": "document",
        "startPageNumber": 1,
        "endPageNumber": 1,
        "unit": "pixel",
        "pages": [
          {
            "pageNumber": 1,
            "spans": []
          }
        ],
        "analyzerId": "prebuilt-imageSearch",
        "mimeType": "image/jpeg"
      }
    ]
  },
  "usage": {
    "contextualizationTokens": 1000,
    "tokens": {
      "gpt-4.1-mini-input": 199,
      "gpt-4.1-mini-output": 106
    }
  }
}

{
  "id": "<request-id>",
  "status": "Succeeded",
  "result": {
    "analyzerId": "prebuilt-audioSearch",
    "apiVersion": "2025-11-01",
    "createdAt": "YYYY-MM-DDTHH:MM:SSZ",
    "stringEncoding": "utf8",
    "warnings": [],
    "contents": [
      {
        "path": "input1",
        "markdown": "# Audio: 00:00.000 => 01:54.670\n\nTranscript\n```\nWEBVTT\n\n00:00.080 --> 00:02.160\n<v Speaker 1>Thank you for calling Woodgrove Travel...",
        "fields": {
          "Summary": {
            "type": "string",
            "valueString": "John Smith contacted Woodgrove Travel to report a negative experience with his flight from New York City to Los Angeles..."
          }
        },
        "kind": "audioVisual",
        "startTimeMs": 0,
        "endTimeMs": 114670,
        "analyzerId": "prebuilt-audioSearch",
        "mimeType": "audio/wav"
      }
    ]
  },
  "usage": {
		"audioHours": 0.032,
    "contextualization": 3194.445
    "tokens": {
      "gpt-4.1-input": 1234, 
      "gpt-4.1-output": 2345,
      "text-embedding-3-large": 3456 
    }
 
	}
}

{
  "id": "2689a699-fa3a-4ddf-9a27-c34ceaa6c597",
  "status": "Succeeded",
  "result": {
    "analyzerId": "prebuilt-videoSearch",
    "apiVersion": "2025-11-01",
    "createdAt": "2025-11-13T16:11:17Z",
    "warnings": [],
    "contents": [
      {
        "markdown": "# Video: 00:00.733 => 00:15.467\nWidth: 1080\nHeight: 608\n\nTranscript\n```\nWEBVTT\n\n00:01.360 --> 00:06.640\n<Speaker 1>When it comes to the neural TTS, in order to get a good voice, it's better to have good data.\n\n00:07.120 --> 00:13.320\n<Speaker 2>To achieve that, we build a universal TTS model based on 3,000 hours of data.\n\n00:13.440 --> 00:23.680\n<Speaker 1>We actually accumulated tons of the data so that this universal model is able to capture the nuance of the audio and generate a more natural voice for the algorithm.\n```\n\nKey Frames\n- 00:00.733 ![](keyFrame.733.jpg)\n- 00:02.067 ![](keyFrame.2067.jpg)\n...\n- 00:14.833 ![](keyFrame.14833.jpg)\n- 00:15.467 ![](keyFrame.15467.jpg)",
        "fields": {
          "Summary": {
            "type": "string",
            "valueString": "The video opens with a Flight Simulator logo alongside Microsoft Azure AI branding, followed by an interview with a man discussing the importance of good data for neural text-to-speech (TTS) technology. He explains the creation of a universal TTS model trained on 3,000 hours of data to capture audio nuances and produce natural-sounding voices. Visuals include audio waveform displays and shots of data centers and server rooms, emphasizing the technological infrastructure behind the TTS model."
          }
        },
        "kind": "audioVisual",
        "startTimeMs": 733,
        "endTimeMs": 15467,
        "width": 1080,
        "height": 608,
        "KeyFrameTimesMs": [
          733,
          2067
          /*... (14 additional keyframes)*/
        ],
        "transcriptPhrases": [
          {
            "speaker": "Speaker 1",
            "startTimeMs": 1360,
            "endTimeMs": 6640,
            "text": "When it comes to the neural TTS, in order to get a good voice, it's better to have good data.",
            "confidence": 0.937,
            "words": [
              {
                "startTimeMs": 1360,
                "endTimeMs": 1600,
                "text": "When"
              },
              {
                "startTimeMs": 1600,
                "endTimeMs": 1760,
                "text": "it"
              }
              /*... (18 additional words)*/
            ],
            "locale": "en-US"
          },
          {
            "speaker": "Speaker 2",
            "startTimeMs": 7120,
            "endTimeMs": 13320,
            "text": "To achieve that, we build a universal TTS model based on 3,000 hours of data.",
            "confidence": 0.937,
            "words": [
              {
                "startTimeMs": 7120,
                "endTimeMs": 7360,
                "text": "To"
              },
              {
                "startTimeMs": 7560,
                "endTimeMs": 7880,
                "text": "achieve"
              }
              /*... (13 additional words)*/
            ],
            "locale": "en-US"
          },
          {
            "speaker": "Speaker 1",
            "startTimeMs": 13440,
            "endTimeMs": 23680,
            "text": "We actually accumulated tons of the data so that this universal model is able to capture the nuance of the audio and generate a more natural voice for the algorithm.",
            "confidence": 0.937,
            "words": [
              {
                "startTimeMs": 13440,
                "endTimeMs": 13600,
                "text": "We"
              },
              {
                "startTimeMs": 13600,
                "endTimeMs": 14000,
                "text": "actually"
              }
              /*... (28 additional words)*/
            ],
            "locale": "en-US"
          }
        ],
        "cameraShotTimesMs": [
          1467,
          3233
          /*... (13 additional camera shots)*/
        ],
        "mimeType": "video/x-m4v"
      },
      {
        "markdown": "# Video: 00:15.467 => 00:23.100\nWidth: 1080\nHeight: 608\n\n\n\nKey Frames\n- 00:15.467 ![](keyFrame.15467.jpg)\n- 00:16.933 ![](keyFrame.16933.jpg)\n...\n- 00:22.367 ![](keyFrame.22367.jpg)\n- 00:23.100 ![](keyFrame.23100.jpg)",
        "fields": {
          "Summary": {
            "type": "string",
            "valueString": "The video transitions to scenic aerial views from the Flight Simulator, showcasing detailed landscapes including coastlines, mountains, and castles. This segment highlights the realistic graphics and immersive experience of the Flight Simulator, demonstrating the integration of advanced AI technologies to enhance the simulation."
          }
        },
        "kind": "audioVisual",
        "startTimeMs": 15467,
        "endTimeMs": 23100,
        "width": 1080,
        "height": 608,
        "KeyFrameTimesMs": [
          15467,
          16933
          /*... (7 additional keyframes)*/
        ],
        "transcriptPhrases": [],
        "cameraShotTimesMs": [
          1467,
          3233
          /*... (13 additional camera shots)*/
        ]
      },
      {
        "markdown": "# Video: 00:23.100 => 00:43.233\nWidth: 1080\nHeight: 608\n\nTranscript\n```\nWEBVTT\n\n00:24.040 --> 00:29.120\n<Speaker 3>What we liked about cognitive services offerings were that they had a much higher fidelity.\n\n00:29.600 --> 00:32.880\n<Speaker 3>And they sounded a lot more like an actual human voice.\n\n00:33.680 --> 00:37.200\n<Speaker 4>Orlando ground 9555 requesting the end of pushback.\n\n00:38.680 --> 00:41.280\n<Speaker 4>9555 request to end pushback received.\n```\n\nKey Frames\n- 00:23.100 ![](keyFrame.23100.jpg)\n- 00:24.833 ![](keyFrame.24833.jpg)\n...\n- 00:42.633 ![](keyFrame.42633.jpg)\n- 00:43.233 ![](keyFrame.43233.jpg)",
        "fields": {
          "Summary": {
            "type": "string",
            "valueString": "The focus shifts back to an interview with another man discussing the high fidelity and human-like quality of voices produced by cognitive services offerings. The segment concludes with visuals of an airplane on the tarmac, ground crew directing the plane, and the plane preparing for pushback, accompanied by realistic audio communications between ground control and the aircraft."
          }
        },
        "kind": "audioVisual",
        "startTimeMs": 23100,
        "endTimeMs": 43233,
        "width": 1080,
        "height": 608,
        "KeyFrameTimesMs": [
          23100,
          24833
          /*... (19 additional keyframes)*/
        ],
        "transcriptPhrases": [
          {
            "speaker": "Speaker 3",
            "startTimeMs": 24040,
            "endTimeMs": 29120,
            "text": "What we liked about cognitive services offerings were that they had a much higher fidelity.",
            "confidence": 0.937,
            "words": [
              {
                "startTimeMs": 24040,
                "endTimeMs": 24240,
                "text": "What"
              },
              {
                "startTimeMs": 24240,
                "endTimeMs": 24320,
                "text": "we"
              }
              /*... (13 additional words)*/
            ],
            "locale": "en-US"
          },
          {
            "speaker": "Speaker 3",
            "startTimeMs": 29600,
            "endTimeMs": 32880,
            "text": "And they sounded a lot more like an actual human voice.",
            "confidence": 0.823,
            "words": [
              {
                "startTimeMs": 29600,
                "endTimeMs": 30080,
                "text": "And"
              },
              {
                "startTimeMs": 30080,
                "endTimeMs": 30160,
                "text": "they"
              }
              /*... (9 additional words)*/
            ],
            "locale": "en-US"
          },
          {
            "speaker": "Speaker 4",
            "startTimeMs": 33680,
            "endTimeMs": 37200,
            "text": "Orlando ground 9555 requesting the end of pushback.",
            "confidence": 0.823,
            "words": [
              {
                "startTimeMs": 33680,
                "endTimeMs": 34160,
                "text": "Orlando"
              },
              {
                "startTimeMs": 34160,
                "endTimeMs": 34600,
                "text": "ground"
              }
              /*... (6 additional words)*/
            ],
            "locale": "en-US"
          },
          {
            "speaker": "Speaker 4",
            "startTimeMs": 38680,
            "endTimeMs": 41280,
            "text": "9555 request to end pushback received.",
            "confidence": 0.823,
            "words": [
              {
                "startTimeMs": 38680,
                "endTimeMs": 39600,
                "text": "9555"
              },
              {
                "startTimeMs": 39600,
                "endTimeMs": 40080,
                "text": "request"
              }
              /*... (4 additional words)*/
            ],
            "locale": "en-US"
          }
        ],
        "cameraShotTimesMs": [
          1467,
          3233
          /*... (13 additional camera shots)*/
        ]
      }
    ]
  },
  "usage": {
    "videoHours": 0.013,
    "contextualizationTokens": 12222,
    "tokens": {
      "gpt-4.1-mini-input": 8976,
      "gpt-4.1-mini-output": 439
    }
  }
}

Próximo passo

Agora que sabe como invocar a operação de análise, saiba mais sobre como construir analisadores personalizados para o seu caso de uso.

Feedback

Esta página foi útil?

Last updated on 2025-12-19