{
  "name": "Spoken Language Identification",
  "description": "Chunk-level language ID over audio based on OpenAI Whisper",
  "app_version": "v0.2",
  "mmif_version": "1.1.0",
  "analyzer_version": "v20250625",
  "app_license": "Apache 2.0",
  "analyzer_license": "MIT",
  "identifier": "http://apps.clams.ai/spoken-lid/v0.2",
  "url": "https://github.com/clamsproject/app-spoken-lid",
  "input": [
    [
      {
        "@type": "http://mmif.clams.ai/vocabulary/AudioDocument/v1",
        "required": true
      },
      {
        "@type": "http://mmif.clams.ai/vocabulary/VideoDocument/v1",
        "required": true
      }
    ]
  ],
  "output": [
    {
      "@type": "http://mmif.clams.ai/vocabulary/TimeFrame/v6",
      "properties": {
        "timeUnit": "seconds",
        "labalSet": "https://raw.githubusercontent.com/openai/whisper/refs/tags/v20250625/whisper/tokenizer.py"
      }
    }
  ],
  "parameters": [
    {
      "name": "model",
      "description": "Whisper model size",
      "type": "string",
      "choices": [
        "tiny",
        "base",
        "small",
        "medium",
        "large",
        "turbo"
      ],
      "default": "tiny",
      "multivalued": false
    },
    {
      "name": "chunk",
      "description": "chunk/window length in seconds",
      "type": "number",
      "default": 30,
      "multivalued": false
    },
    {
      "name": "top",
      "description": "top-k language scores",
      "type": "integer",
      "default": 3,
      "multivalued": false
    },
    {
      "name": "batchSize",
      "description": "number of windows processed in a batch",
      "type": "integer",
      "default": 1,
      "multivalued": false
    },
    {
      "name": "pretty",
      "description": "The JSON body of the HTTP response will be re-formatted with 2-space indentation",
      "type": "boolean",
      "default": false,
      "multivalued": false
    },
    {
      "name": "runningTime",
      "description": "The running time of the app will be recorded in the view metadata",
      "type": "boolean",
      "default": false,
      "multivalued": false
    },
    {
      "name": "hwFetch",
      "description": "The hardware information (architecture, GPU and vRAM) will be recorded in the view metadata",
      "type": "boolean",
      "default": false,
      "multivalued": false
    }
  ]
}