Transcription streaming events

For the complete documentation index, see llms.txt. Markdown versions of documentation pages are available by appending .md to the page URL.

transcript.text.segment

Emitted when a diarized transcription returns a completed segment with speaker information. Only emitted when you create a transcription with stream set to true and response_format set to diarized_json.

Schema

Schema name: TranscriptTextSegmentEvent

{
  "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema)": {
    "kind": "HttpDeclTypeAlias",
    "oasRef": "#/components/schemas/TranscriptTextSegmentEvent",
    "docstring": "Emitted when a diarized transcription returns a completed segment with speaker information. Only emitted when you [create a transcription](/docs/api-reference/audio/create-transcription) with `stream` set to `true` and `response_format` set to `diarized_json`.\n",
    "ident": "TranscriptionTextSegmentEvent",
    "type": {
      "kind": "HttpTypeObject",
      "members": [
        {
          "ident": "id"
        },
        {
          "ident": "end"
        },
        {
          "ident": "speaker"
        },
        {
          "ident": "start"
        },
        {
          "ident": "text"
        },
        {
          "ident": "type"
        }
      ]
    },
    "childrenParentSchema": "object",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) id",
      "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) end",
      "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) speaker",
      "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) start",
      "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) text",
      "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) type"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) id": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/id",
    "deprecated": false,
    "key": "id",
    "docstring": "Unique identifier for the segment.",
    "type": {
      "kind": "HttpTypeString"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "string",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) end": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/end",
    "deprecated": false,
    "key": "end",
    "docstring": "End timestamp of the segment in seconds.",
    "type": {
      "kind": "HttpTypeNumber"
    },
    "constraints": {
      "format": "double"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "number",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) speaker": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/speaker",
    "deprecated": false,
    "key": "speaker",
    "docstring": "Speaker label for this segment.",
    "type": {
      "kind": "HttpTypeString"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "string",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) start": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/start",
    "deprecated": false,
    "key": "start",
    "docstring": "Start timestamp of the segment in seconds.",
    "type": {
      "kind": "HttpTypeNumber"
    },
    "constraints": {
      "format": "double"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "number",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) text": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/text",
    "deprecated": false,
    "key": "text",
    "docstring": "Transcript text for this segment.",
    "type": {
      "kind": "HttpTypeString"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "string",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) type": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/type",
    "deprecated": false,
    "key": "type",
    "docstring": "The type of the event. Always `transcript.text.segment`.",
    "type": {
      "kind": "HttpTypeUnion",
      "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/type",
      "types": [
        {
          "kind": "HttpTypeLiteral",
          "literal": "transcript.text.segment"
        }
      ]
    },
    "optional": false,
    "nullable": false,
    "schemaType": "enum",
    "childrenParentSchema": "enum",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) type > (member) 0"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) type > (member) 0": {
    "kind": "HttpDeclReference",
    "type": {
      "kind": "HttpTypeLiteral",
      "literal": "transcript.text.segment"
    }
  }
}

Example

{
  "type": "transcript.text.segment",
  "id": "seg_002",
  "start": 5.2,
  "end": 12.8,
  "text": "Hi, I need help with diarization.",
  "speaker": "A"
}

transcript.text.delta

Emitted when there is an additional text delta. This is also the first event emitted when the transcription starts. Only emitted when you create a transcription with the Stream parameter set to true.

Schema

Schema name: TranscriptTextDeltaEvent

{
  "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema)": {
    "kind": "HttpDeclTypeAlias",
    "oasRef": "#/components/schemas/TranscriptTextDeltaEvent",
    "docstring": "Emitted when there is an additional text delta. This is also the first event emitted when the transcription starts. Only emitted when you [create a transcription](/docs/api-reference/audio/create-transcription) with the `Stream` parameter set to `true`.",
    "ident": "TranscriptionTextDeltaEvent",
    "type": {
      "kind": "HttpTypeObject",
      "members": [
        {
          "ident": "delta"
        },
        {
          "ident": "type"
        },
        {
          "ident": "logprobs"
        },
        {
          "ident": "segment_id"
        }
      ]
    },
    "childrenParentSchema": "object",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) delta",
      "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) type",
      "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs",
      "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) segment_id"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) delta": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/delta",
    "deprecated": false,
    "key": "delta",
    "docstring": "The text delta that was additionally transcribed.\n",
    "type": {
      "kind": "HttpTypeString"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "string",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) type": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/type",
    "deprecated": false,
    "key": "type",
    "docstring": "The type of the event. Always `transcript.text.delta`.\n",
    "type": {
      "kind": "HttpTypeUnion",
      "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/type",
      "types": [
        {
          "kind": "HttpTypeLiteral",
          "literal": "transcript.text.delta"
        }
      ]
    },
    "optional": false,
    "nullable": false,
    "schemaType": "enum",
    "childrenParentSchema": "enum",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) type > (member) 0"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs",
    "deprecated": false,
    "key": "logprobs",
    "docstring": "The log probabilities of the delta. Only included if you [create a transcription](/docs/api-reference/audio/create-transcription) with the `include[]` parameter set to `logprobs`.\n",
    "type": {
      "kind": "HttpTypeArray",
      "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs",
      "elementType": {
        "kind": "HttpTypeObject",
        "members": [
          {
            "ident": "token"
          },
          {
            "ident": "bytes"
          },
          {
            "ident": "logprob"
          }
        ]
      }
    },
    "optional": true,
    "nullable": false,
    "schemaType": "array",
    "childrenParentSchema": "object",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) token",
      "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) bytes",
      "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) logprob"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) segment_id": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/segment_id",
    "deprecated": false,
    "key": "segment_id",
    "docstring": "Identifier of the diarized segment that this delta belongs to. Only present when using `gpt-4o-transcribe-diarize`.\n",
    "type": {
      "kind": "HttpTypeString"
    },
    "optional": true,
    "nullable": false,
    "schemaType": "string",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) type > (member) 0": {
    "kind": "HttpDeclReference",
    "type": {
      "kind": "HttpTypeLiteral",
      "literal": "transcript.text.delta"
    }
  },
  "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) token": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs/items/properties/token",
    "deprecated": false,
    "key": "token",
    "docstring": "The token that was used to generate the log probability.\n",
    "type": {
      "kind": "HttpTypeString"
    },
    "optional": true,
    "nullable": false,
    "schemaType": "string",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) bytes": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs/items/properties/bytes",
    "deprecated": false,
    "key": "bytes",
    "docstring": "The bytes that were used to generate the log probability.\n",
    "type": {
      "kind": "HttpTypeArray",
      "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs/items/properties/bytes",
      "elementType": {
        "kind": "HttpTypeNumber"
      }
    },
    "optional": true,
    "nullable": false,
    "schemaType": "array",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) logprob": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs/items/properties/logprob",
    "deprecated": false,
    "key": "logprob",
    "docstring": "The log probability of the token.\n",
    "type": {
      "kind": "HttpTypeNumber"
    },
    "optional": true,
    "nullable": false,
    "schemaType": "number",
    "children": []
  }
}

Example

{
  "type": "transcript.text.delta",
  "delta": " wonderful"
}

transcript.text.done

Emitted when the transcription is complete. Contains the complete transcription text. Only emitted when you create a transcription with the Stream parameter set to true.

Schema

Schema name: TranscriptTextDoneEvent

{
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema)": {
    "kind": "HttpDeclTypeAlias",
    "oasRef": "#/components/schemas/TranscriptTextDoneEvent",
    "docstring": "Emitted when the transcription is complete. Contains the complete transcription text. Only emitted when you [create a transcription](/docs/api-reference/audio/create-transcription) with the `Stream` parameter set to `true`.",
    "ident": "TranscriptionTextDoneEvent",
    "type": {
      "kind": "HttpTypeObject",
      "members": [
        {
          "ident": "text"
        },
        {
          "ident": "type"
        },
        {
          "ident": "languages"
        },
        {
          "ident": "logprobs"
        },
        {
          "ident": "usage"
        }
      ]
    },
    "childrenParentSchema": "object",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) text",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) type",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) languages",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) text": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/text",
    "deprecated": false,
    "key": "text",
    "docstring": "The text that was transcribed.\n",
    "type": {
      "kind": "HttpTypeString"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "string",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) type": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/type",
    "deprecated": false,
    "key": "type",
    "docstring": "The type of the event. Always `transcript.text.done`.\n",
    "type": {
      "kind": "HttpTypeUnion",
      "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/type",
      "types": [
        {
          "kind": "HttpTypeLiteral",
          "literal": "transcript.text.done"
        }
      ]
    },
    "optional": false,
    "nullable": false,
    "schemaType": "enum",
    "childrenParentSchema": "enum",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) type > (member) 0"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) languages": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/languages",
    "deprecated": false,
    "key": "languages",
    "docstring": "The languages detected in the audio. Returned by `gpt-transcribe`. An empty array indicates that no language could be reliably detected.\n",
    "type": {
      "kind": "HttpTypeArray",
      "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/languages",
      "elementType": {
        "kind": "HttpTypeReference",
        "ident": "TranscriptionLanguage",
        "$ref": "(resource) audio.transcriptions > (model) transcription_language > (schema)"
      }
    },
    "optional": true,
    "nullable": false,
    "schemaType": "array",
    "childrenParentSchema": "object",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_language > (schema) > (property) code"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs",
    "deprecated": false,
    "key": "logprobs",
    "docstring": "The log probabilities of the individual tokens in the transcription. Only included if you [create a transcription](/docs/api-reference/audio/create-transcription) with the `include[]` parameter set to `logprobs`.\n",
    "type": {
      "kind": "HttpTypeArray",
      "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs",
      "elementType": {
        "kind": "HttpTypeObject",
        "members": [
          {
            "ident": "token"
          },
          {
            "ident": "bytes"
          },
          {
            "ident": "logprob"
          }
        ]
      }
    },
    "optional": true,
    "nullable": false,
    "schemaType": "array",
    "childrenParentSchema": "object",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) token",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) bytes",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) logprob"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/usage",
    "deprecated": false,
    "key": "usage",
    "docstring": "Usage statistics for models billed by token usage.",
    "title": "Token Usage",
    "type": {
      "kind": "HttpTypeObject",
      "members": [
        {
          "ident": "input_tokens"
        },
        {
          "ident": "output_tokens"
        },
        {
          "ident": "total_tokens"
        },
        {
          "ident": "type"
        },
        {
          "ident": "input_token_details"
        }
      ]
    },
    "optional": true,
    "nullable": false,
    "schemaType": "object",
    "childrenParentSchema": "object",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_tokens",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) output_tokens",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) total_tokens",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) type",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) type > (member) 0": {
    "kind": "HttpDeclReference",
    "type": {
      "kind": "HttpTypeLiteral",
      "literal": "transcript.text.done"
    }
  },
  "(resource) audio.transcriptions > (model) transcription_language > (schema) > (property) code": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptionLanguage/properties/code",
    "deprecated": false,
    "key": "code",
    "docstring": "The code of a language detected in the audio.",
    "type": {
      "kind": "HttpTypeString"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "string",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_language > (schema)": {
    "kind": "HttpDeclTypeAlias",
    "oasRef": "#/components/schemas/TranscriptionLanguage",
    "docstring": "A language detected in transcribed audio.",
    "ident": "TranscriptionLanguage",
    "type": {
      "kind": "HttpTypeObject",
      "members": [
        {
          "ident": "code"
        }
      ]
    },
    "childrenParentSchema": "object",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_language > (schema) > (property) code"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) token": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs/items/properties/token",
    "deprecated": false,
    "key": "token",
    "docstring": "The token that was used to generate the log probability.\n",
    "type": {
      "kind": "HttpTypeString"
    },
    "optional": true,
    "nullable": false,
    "schemaType": "string",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) bytes": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs/items/properties/bytes",
    "deprecated": false,
    "key": "bytes",
    "docstring": "The bytes that were used to generate the log probability.\n",
    "type": {
      "kind": "HttpTypeArray",
      "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs/items/properties/bytes",
      "elementType": {
        "kind": "HttpTypeNumber"
      }
    },
    "optional": true,
    "nullable": false,
    "schemaType": "array",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) logprob": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs/items/properties/logprob",
    "deprecated": false,
    "key": "logprob",
    "docstring": "The log probability of the token.\n",
    "type": {
      "kind": "HttpTypeNumber"
    },
    "optional": true,
    "nullable": false,
    "schemaType": "number",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_tokens": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/input_tokens",
    "deprecated": false,
    "key": "input_tokens",
    "docstring": "Number of input tokens billed for this request.",
    "type": {
      "kind": "HttpTypeNumber"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "integer",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) output_tokens": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/output_tokens",
    "deprecated": false,
    "key": "output_tokens",
    "docstring": "Number of output tokens generated.",
    "type": {
      "kind": "HttpTypeNumber"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "integer",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) total_tokens": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/total_tokens",
    "deprecated": false,
    "key": "total_tokens",
    "docstring": "Total number of tokens used (input + output).",
    "type": {
      "kind": "HttpTypeNumber"
    },
    "optional": false,
    "nullable": false,
    "schemaType": "integer",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) type": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/type",
    "deprecated": false,
    "key": "type",
    "docstring": "The type of the usage object. Always `tokens` for this variant.",
    "type": {
      "kind": "HttpTypeUnion",
      "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/type",
      "types": [
        {
          "kind": "HttpTypeLiteral",
          "literal": "tokens"
        }
      ]
    },
    "optional": false,
    "nullable": false,
    "schemaType": "enum",
    "childrenParentSchema": "enum",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) type > (member) 0"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/input_token_details",
    "deprecated": false,
    "key": "input_token_details",
    "docstring": "Details about the input tokens billed for this request.",
    "type": {
      "kind": "HttpTypeObject",
      "members": [
        {
          "ident": "audio_tokens"
        },
        {
          "ident": "text_tokens"
        }
      ]
    },
    "optional": true,
    "nullable": false,
    "schemaType": "object",
    "childrenParentSchema": "object",
    "children": [
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details > (property) audio_tokens",
      "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details > (property) text_tokens"
    ]
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) type > (member) 0": {
    "kind": "HttpDeclReference",
    "type": {
      "kind": "HttpTypeLiteral",
      "literal": "tokens"
    }
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details > (property) audio_tokens": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/input_token_details/properties/audio_tokens",
    "deprecated": false,
    "key": "audio_tokens",
    "docstring": "Number of audio tokens billed for this request.",
    "type": {
      "kind": "HttpTypeNumber"
    },
    "optional": true,
    "nullable": false,
    "schemaType": "integer",
    "children": []
  },
  "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details > (property) text_tokens": {
    "kind": "HttpDeclProperty",
    "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/input_token_details/properties/text_tokens",
    "deprecated": false,
    "key": "text_tokens",
    "docstring": "Number of text tokens billed for this request.",
    "type": {
      "kind": "HttpTypeNumber"
    },
    "optional": true,
    "nullable": false,
    "schemaType": "integer",
    "children": []
  }
}

Example

{
  "type": "transcript.text.done",
  "text": "I see skies of blue and clouds of white, the bright blessed days, the dark sacred nights, and I think to myself, what a wonderful world.",
  "usage": {
    "type": "tokens",
    "input_tokens": 14,
    "input_token_details": {
      "text_tokens": 10,
      "audio_tokens": 4
    },
    "output_tokens": 31,
    "total_tokens": 45
  }
}

resources/audio/subresources/transcriptions/streaming-events.md +777 −0 created

1# Transcription streaming events

3> For the complete documentation index, see [llms.txt](/llms.txt). Markdown versions of documentation pages are available by appending `.md` to the page URL.

5## transcript.text.segment

7Emitted when a diarized transcription returns a completed segment with speaker information. Only emitted when you [create a transcription](https://developers.openai.com/docs/api-reference/audio/create-transcription) with `stream` set to `true` and `response_format` set to `diarized_json`.

9### Schema

11Schema name: `TranscriptTextSegmentEvent`

13```json

14{

15 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema)": {

16 "kind": "HttpDeclTypeAlias",

17 "oasRef": "#/components/schemas/TranscriptTextSegmentEvent",

18 "docstring": "Emitted when a diarized transcription returns a completed segment with speaker information. Only emitted when you [create a transcription](/docs/api-reference/audio/create-transcription) with `stream` set to `true` and `response_format` set to `diarized_json`.\n",

19 "ident": "TranscriptionTextSegmentEvent",

20 "type": {

21 "kind": "HttpTypeObject",

22 "members": [

23 {

24 "ident": "id"

25 },

26 {

27 "ident": "end"

28 },

29 {

30 "ident": "speaker"

31 },

32 {

33 "ident": "start"

34 },

35 {

36 "ident": "text"

37 },

38 {

39 "ident": "type"

40 }

41 ]

42 },

43 "childrenParentSchema": "object",

44 "children": [

45 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) id",

46 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) end",

47 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) speaker",

48 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) start",

49 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) text",

50 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) type"

51 ]

52 },

53 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) id": {

54 "kind": "HttpDeclProperty",

55 "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/id",

56 "deprecated": false,

57 "key": "id",

58 "docstring": "Unique identifier for the segment.",

59 "type": {

60 "kind": "HttpTypeString"

61 },

62 "optional": false,

63 "nullable": false,

64 "schemaType": "string",

65 "children": []

66 },

67 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) end": {

68 "kind": "HttpDeclProperty",

69 "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/end",

70 "deprecated": false,

71 "key": "end",

72 "docstring": "End timestamp of the segment in seconds.",

73 "type": {

74 "kind": "HttpTypeNumber"

75 },

76 "constraints": {

77 "format": "double"

78 },

79 "optional": false,

80 "nullable": false,

81 "schemaType": "number",

82 "children": []

83 },

84 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) speaker": {

85 "kind": "HttpDeclProperty",

86 "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/speaker",

87 "deprecated": false,

88 "key": "speaker",

89 "docstring": "Speaker label for this segment.",

90 "type": {

91 "kind": "HttpTypeString"

92 },

93 "optional": false,

94 "nullable": false,

95 "schemaType": "string",

96 "children": []

97 },

98 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) start": {

99 "kind": "HttpDeclProperty",

100 "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/start",

101 "deprecated": false,

102 "key": "start",

103 "docstring": "Start timestamp of the segment in seconds.",

104 "type": {

105 "kind": "HttpTypeNumber"

106 },

107 "constraints": {

108 "format": "double"

109 },

110 "optional": false,

111 "nullable": false,

112 "schemaType": "number",

113 "children": []

114 },

115 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) text": {

116 "kind": "HttpDeclProperty",

117 "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/text",

118 "deprecated": false,

119 "key": "text",

120 "docstring": "Transcript text for this segment.",

121 "type": {

122 "kind": "HttpTypeString"

123 },

124 "optional": false,

125 "nullable": false,

126 "schemaType": "string",

127 "children": []

128 },

129 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) type": {

130 "kind": "HttpDeclProperty",

131 "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/type",

132 "deprecated": false,

133 "key": "type",

134 "docstring": "The type of the event. Always `transcript.text.segment`.",

135 "type": {

136 "kind": "HttpTypeUnion",

137 "oasRef": "#/components/schemas/TranscriptTextSegmentEvent/properties/type",

138 "types": [

139 {

140 "kind": "HttpTypeLiteral",

141 "literal": "transcript.text.segment"

142 }

143 ]

144 },

145 "optional": false,

146 "nullable": false,

147 "schemaType": "enum",

148 "childrenParentSchema": "enum",

149 "children": [

150 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) type > (member) 0"

151 ]

152 },

153 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema) > (property) type > (member) 0": {

154 "kind": "HttpDeclReference",

155 "type": {

156 "kind": "HttpTypeLiteral",

157 "literal": "transcript.text.segment"

158 }

159 }

160}

161```

162

163### Example

164

165```json

166{

167 "type": "transcript.text.segment",

168 "id": "seg_002",

169 "start": 5.2,

170 "end": 12.8,

171 "text": "Hi, I need help with diarization.",

172 "speaker": "A"

173}

174```

175

176## transcript.text.delta

177

178Emitted when there is an additional text delta. This is also the first event emitted when the transcription starts. Only emitted when you [create a transcription](https://developers.openai.com/docs/api-reference/audio/create-transcription) with the `Stream` parameter set to `true`.

179

180### Schema

181

182Schema name: `TranscriptTextDeltaEvent`

183

184```json

185{

186 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema)": {

187 "kind": "HttpDeclTypeAlias",

188 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent",

189 "docstring": "Emitted when there is an additional text delta. This is also the first event emitted when the transcription starts. Only emitted when you [create a transcription](/docs/api-reference/audio/create-transcription) with the `Stream` parameter set to `true`.",

190 "ident": "TranscriptionTextDeltaEvent",

191 "type": {

192 "kind": "HttpTypeObject",

193 "members": [

194 {

195 "ident": "delta"

196 },

197 {

198 "ident": "type"

199 },

200 {

201 "ident": "logprobs"

202 },

203 {

204 "ident": "segment_id"

205 }

206 ]

207 },

208 "childrenParentSchema": "object",

209 "children": [

210 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) delta",

211 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) type",

212 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs",

213 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) segment_id"

214 ]

215 },

216 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) delta": {

217 "kind": "HttpDeclProperty",

218 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/delta",

219 "deprecated": false,

220 "key": "delta",

221 "docstring": "The text delta that was additionally transcribed.\n",

222 "type": {

223 "kind": "HttpTypeString"

224 },

225 "optional": false,

226 "nullable": false,

227 "schemaType": "string",

228 "children": []

229 },

230 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) type": {

231 "kind": "HttpDeclProperty",

232 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/type",

233 "deprecated": false,

234 "key": "type",

235 "docstring": "The type of the event. Always `transcript.text.delta`.\n",

236 "type": {

237 "kind": "HttpTypeUnion",

238 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/type",

239 "types": [

240 {

241 "kind": "HttpTypeLiteral",

242 "literal": "transcript.text.delta"

243 }

244 ]

245 },

246 "optional": false,

247 "nullable": false,

248 "schemaType": "enum",

249 "childrenParentSchema": "enum",

250 "children": [

251 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) type > (member) 0"

252 ]

253 },

254 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs": {

255 "kind": "HttpDeclProperty",

256 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs",

257 "deprecated": false,

258 "key": "logprobs",

259 "docstring": "The log probabilities of the delta. Only included if you [create a transcription](/docs/api-reference/audio/create-transcription) with the `include[]` parameter set to `logprobs`.\n",

260 "type": {

261 "kind": "HttpTypeArray",

262 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs",

263 "elementType": {

264 "kind": "HttpTypeObject",

265 "members": [

266 {

267 "ident": "token"

268 },

269 {

270 "ident": "bytes"

271 },

272 {

273 "ident": "logprob"

274 }

275 ]

276 }

277 },

278 "optional": true,

279 "nullable": false,

280 "schemaType": "array",

281 "childrenParentSchema": "object",

282 "children": [

283 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) token",

284 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) bytes",

285 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) logprob"

286 ]

287 },

288 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) segment_id": {

289 "kind": "HttpDeclProperty",

290 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/segment_id",

291 "deprecated": false,

292 "key": "segment_id",

293 "docstring": "Identifier of the diarized segment that this delta belongs to. Only present when using `gpt-4o-transcribe-diarize`.\n",

294 "type": {

295 "kind": "HttpTypeString"

296 },

297 "optional": true,

298 "nullable": false,

299 "schemaType": "string",

300 "children": []

301 },

302 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) type > (member) 0": {

303 "kind": "HttpDeclReference",

304 "type": {

305 "kind": "HttpTypeLiteral",

306 "literal": "transcript.text.delta"

307 }

308 },

309 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) token": {

310 "kind": "HttpDeclProperty",

311 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs/items/properties/token",

312 "deprecated": false,

313 "key": "token",

314 "docstring": "The token that was used to generate the log probability.\n",

315 "type": {

316 "kind": "HttpTypeString"

317 },

318 "optional": true,

319 "nullable": false,

320 "schemaType": "string",

321 "children": []

322 },

323 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) bytes": {

324 "kind": "HttpDeclProperty",

325 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs/items/properties/bytes",

326 "deprecated": false,

327 "key": "bytes",

328 "docstring": "The bytes that were used to generate the log probability.\n",

329 "type": {

330 "kind": "HttpTypeArray",

331 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs/items/properties/bytes",

332 "elementType": {

333 "kind": "HttpTypeNumber"

334 }

335 },

336 "optional": true,

337 "nullable": false,

338 "schemaType": "array",

339 "children": []

340 },

341 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema) > (property) logprobs > (items) > (property) logprob": {

342 "kind": "HttpDeclProperty",

343 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs/items/properties/logprob",

344 "deprecated": false,

345 "key": "logprob",

346 "docstring": "The log probability of the token.\n",

347 "type": {

348 "kind": "HttpTypeNumber"

349 },

350 "optional": true,

351 "nullable": false,

352 "schemaType": "number",

353 "children": []

354 }

355}

356```

357

358### Example

359

360```json

361{

362 "type": "transcript.text.delta",

363 "delta": " wonderful"

364}

365```

366

367## transcript.text.done

368

369Emitted when the transcription is complete. Contains the complete transcription text. Only emitted when you [create a transcription](https://developers.openai.com/docs/api-reference/audio/create-transcription) with the `Stream` parameter set to `true`.

370

371### Schema

372

373Schema name: `TranscriptTextDoneEvent`

374

375```json

376{

377 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema)": {

378 "kind": "HttpDeclTypeAlias",

379 "oasRef": "#/components/schemas/TranscriptTextDoneEvent",

380 "docstring": "Emitted when the transcription is complete. Contains the complete transcription text. Only emitted when you [create a transcription](/docs/api-reference/audio/create-transcription) with the `Stream` parameter set to `true`.",

381 "ident": "TranscriptionTextDoneEvent",

382 "type": {

383 "kind": "HttpTypeObject",

384 "members": [

385 {

386 "ident": "text"

387 },

388 {

389 "ident": "type"

390 },

391 {

392 "ident": "languages"

393 },

394 {

395 "ident": "logprobs"

396 },

397 {

398 "ident": "usage"

399 }

400 ]

401 },

402 "childrenParentSchema": "object",

403 "children": [

404 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) text",

405 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) type",

406 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) languages",

407 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs",

408 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage"

409 ]

410 },

411 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) text": {

412 "kind": "HttpDeclProperty",

413 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/text",

414 "deprecated": false,

415 "key": "text",

416 "docstring": "The text that was transcribed.\n",

417 "type": {

418 "kind": "HttpTypeString"

419 },

420 "optional": false,

421 "nullable": false,

422 "schemaType": "string",

423 "children": []

424 },

425 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) type": {

426 "kind": "HttpDeclProperty",

427 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/type",

428 "deprecated": false,

429 "key": "type",

430 "docstring": "The type of the event. Always `transcript.text.done`.\n",

431 "type": {

432 "kind": "HttpTypeUnion",

433 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/type",

434 "types": [

435 {

436 "kind": "HttpTypeLiteral",

437 "literal": "transcript.text.done"

438 }

439 ]

440 },

441 "optional": false,

442 "nullable": false,

443 "schemaType": "enum",

444 "childrenParentSchema": "enum",

445 "children": [

446 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) type > (member) 0"

447 ]

448 },

449 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) languages": {

450 "kind": "HttpDeclProperty",

451 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/languages",

452 "deprecated": false,

453 "key": "languages",

454 "docstring": "The languages detected in the audio. Returned by `gpt-transcribe`. An empty array indicates that no language could be reliably detected.\n",

455 "type": {

456 "kind": "HttpTypeArray",

457 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/languages",

458 "elementType": {

459 "kind": "HttpTypeReference",

460 "ident": "TranscriptionLanguage",

461 "$ref": "(resource) audio.transcriptions > (model) transcription_language > (schema)"

462 }

463 },

464 "optional": true,

465 "nullable": false,

466 "schemaType": "array",

467 "childrenParentSchema": "object",

468 "children": [

469 "(resource) audio.transcriptions > (model) transcription_language > (schema) > (property) code"

470 ]

471 },

472 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs": {

473 "kind": "HttpDeclProperty",

474 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs",

475 "deprecated": false,

476 "key": "logprobs",

477 "docstring": "The log probabilities of the individual tokens in the transcription. Only included if you [create a transcription](/docs/api-reference/audio/create-transcription) with the `include[]` parameter set to `logprobs`.\n",

478 "type": {

479 "kind": "HttpTypeArray",

480 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs",

481 "elementType": {

482 "kind": "HttpTypeObject",

483 "members": [

484 {

485 "ident": "token"

486 },

487 {

488 "ident": "bytes"

489 },

490 {

491 "ident": "logprob"

492 }

493 ]

494 }

495 },

496 "optional": true,

497 "nullable": false,

498 "schemaType": "array",

499 "childrenParentSchema": "object",

500 "children": [

501 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) token",

502 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) bytes",

503 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) logprob"

504 ]

505 },

506 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage": {

507 "kind": "HttpDeclProperty",

508 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/usage",

509 "deprecated": false,

510 "key": "usage",

511 "docstring": "Usage statistics for models billed by token usage.",

512 "title": "Token Usage",

513 "type": {

514 "kind": "HttpTypeObject",

515 "members": [

516 {

517 "ident": "input_tokens"

518 },

519 {

520 "ident": "output_tokens"

521 },

522 {

523 "ident": "total_tokens"

524 },

525 {

526 "ident": "type"

527 },

528 {

529 "ident": "input_token_details"

530 }

531 ]

532 },

533 "optional": true,

534 "nullable": false,

535 "schemaType": "object",

536 "childrenParentSchema": "object",

537 "children": [

538 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_tokens",

539 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) output_tokens",

540 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) total_tokens",

541 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) type",

542 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details"

543 ]

544 },

545 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) type > (member) 0": {

546 "kind": "HttpDeclReference",

547 "type": {

548 "kind": "HttpTypeLiteral",

549 "literal": "transcript.text.done"

550 }

551 },

552 "(resource) audio.transcriptions > (model) transcription_language > (schema) > (property) code": {

553 "kind": "HttpDeclProperty",

554 "oasRef": "#/components/schemas/TranscriptionLanguage/properties/code",

555 "deprecated": false,

556 "key": "code",

557 "docstring": "The code of a language detected in the audio.",

558 "type": {

559 "kind": "HttpTypeString"

560 },

561 "optional": false,

562 "nullable": false,

563 "schemaType": "string",

564 "children": []

565 },

566 "(resource) audio.transcriptions > (model) transcription_language > (schema)": {

567 "kind": "HttpDeclTypeAlias",

568 "oasRef": "#/components/schemas/TranscriptionLanguage",

569 "docstring": "A language detected in transcribed audio.",

570 "ident": "TranscriptionLanguage",

571 "type": {

572 "kind": "HttpTypeObject",

573 "members": [

574 {

575 "ident": "code"

576 }

577 ]

578 },

579 "childrenParentSchema": "object",

580 "children": [

581 "(resource) audio.transcriptions > (model) transcription_language > (schema) > (property) code"

582 ]

583 },

584 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) token": {

585 "kind": "HttpDeclProperty",

586 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs/items/properties/token",

587 "deprecated": false,

588 "key": "token",

589 "docstring": "The token that was used to generate the log probability.\n",

590 "type": {

591 "kind": "HttpTypeString"

592 },

593 "optional": true,

594 "nullable": false,

595 "schemaType": "string",

596 "children": []

597 },

598 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) bytes": {

599 "kind": "HttpDeclProperty",

600 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs/items/properties/bytes",

601 "deprecated": false,

602 "key": "bytes",

603 "docstring": "The bytes that were used to generate the log probability.\n",

604 "type": {

605 "kind": "HttpTypeArray",

606 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs/items/properties/bytes",

607 "elementType": {

608 "kind": "HttpTypeNumber"

609 }

610 },

611 "optional": true,

612 "nullable": false,

613 "schemaType": "array",

614 "children": []

615 },

616 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) logprobs > (items) > (property) logprob": {

617 "kind": "HttpDeclProperty",

618 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs/items/properties/logprob",

619 "deprecated": false,

620 "key": "logprob",

621 "docstring": "The log probability of the token.\n",

622 "type": {

623 "kind": "HttpTypeNumber"

624 },

625 "optional": true,

626 "nullable": false,

627 "schemaType": "number",

628 "children": []

629 },

630 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_tokens": {

631 "kind": "HttpDeclProperty",

632 "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/input_tokens",

633 "deprecated": false,

634 "key": "input_tokens",

635 "docstring": "Number of input tokens billed for this request.",

636 "type": {

637 "kind": "HttpTypeNumber"

638 },

639 "optional": false,

640 "nullable": false,

641 "schemaType": "integer",

642 "children": []

643 },

644 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) output_tokens": {

645 "kind": "HttpDeclProperty",

646 "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/output_tokens",

647 "deprecated": false,

648 "key": "output_tokens",

649 "docstring": "Number of output tokens generated.",

650 "type": {

651 "kind": "HttpTypeNumber"

652 },

653 "optional": false,

654 "nullable": false,

655 "schemaType": "integer",

656 "children": []

657 },

658 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) total_tokens": {

659 "kind": "HttpDeclProperty",

660 "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/total_tokens",

661 "deprecated": false,

662 "key": "total_tokens",

663 "docstring": "Total number of tokens used (input + output).",

664 "type": {

665 "kind": "HttpTypeNumber"

666 },

667 "optional": false,

668 "nullable": false,

669 "schemaType": "integer",

670 "children": []

671 },

672 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) type": {

673 "kind": "HttpDeclProperty",

674 "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/type",

675 "deprecated": false,

676 "key": "type",

677 "docstring": "The type of the usage object. Always `tokens` for this variant.",

678 "type": {

679 "kind": "HttpTypeUnion",

680 "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/type",

681 "types": [

682 {

683 "kind": "HttpTypeLiteral",

684 "literal": "tokens"

685 }

686 ]

687 },

688 "optional": false,

689 "nullable": false,

690 "schemaType": "enum",

691 "childrenParentSchema": "enum",

692 "children": [

693 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) type > (member) 0"

694 ]

695 },

696 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details": {

697 "kind": "HttpDeclProperty",

698 "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/input_token_details",

699 "deprecated": false,

700 "key": "input_token_details",

701 "docstring": "Details about the input tokens billed for this request.",

702 "type": {

703 "kind": "HttpTypeObject",

704 "members": [

705 {

706 "ident": "audio_tokens"

707 },

708 {

709 "ident": "text_tokens"

710 }

711 ]

712 },

713 "optional": true,

714 "nullable": false,

715 "schemaType": "object",

716 "childrenParentSchema": "object",

717 "children": [

718 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details > (property) audio_tokens",

719 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details > (property) text_tokens"

720 ]

721 },

722 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) type > (member) 0": {

723 "kind": "HttpDeclReference",

724 "type": {

725 "kind": "HttpTypeLiteral",

726 "literal": "tokens"

727 }

728 },

729 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details > (property) audio_tokens": {

730 "kind": "HttpDeclProperty",

731 "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/input_token_details/properties/audio_tokens",

732 "deprecated": false,

733 "key": "audio_tokens",

734 "docstring": "Number of audio tokens billed for this request.",

735 "type": {

736 "kind": "HttpTypeNumber"

737 },

738 "optional": true,

739 "nullable": false,

740 "schemaType": "integer",

741 "children": []

742 },

743 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema) > (property) usage > (property) input_token_details > (property) text_tokens": {

744 "kind": "HttpDeclProperty",

745 "oasRef": "#/components/schemas/TranscriptTextUsageTokens/properties/input_token_details/properties/text_tokens",

746 "deprecated": false,

747 "key": "text_tokens",

748 "docstring": "Number of text tokens billed for this request.",

749 "type": {

750 "kind": "HttpTypeNumber"

751 },

752 "optional": true,

753 "nullable": false,

754 "schemaType": "integer",

755 "children": []

756 }

757}

758```

759

760### Example

761

762```json

763{

764 "type": "transcript.text.done",

765 "text": "I see skies of blue and clouds of white, the bright blessed days, the dark sacred nights, and I think to myself, what a wonderful world.",

766 "usage": {

767 "type": "tokens",

768 "input_tokens": 14,

769 "input_token_details": {

770 "text_tokens": 10,

771 "audio_tokens": 4

772 },

773 "output_tokens": 31,

774 "total_tokens": 45

775 }

776}

777```

resources/audio/subresources/transcriptions/streaming-events.md 2026-09-03 23:00 UTC to 2026-09-04 23:59 UTC

Transcription streaming events

transcript.text.segment

Schema

Example

transcript.text.delta

Schema

Example

transcript.text.done

Schema

Example