Server-side events

This topic describes the server-side events for the Qwen-TTS-Realtime API.

For more information, see Real-time speech synthesis - Qwen.

error

The server returns this event when it encounters a client-side or server-side error.

Parameter	Type	Description
type	string	The event type. The value is fixed to `error`.
error	object	The detailed information about the error.
error.type	string	The error type.
error.code	string	The error code.
error.message	string	The error message.
error.param	string	The parameter related to the error, such as `session.voice`.

{
  "event_id": "event_B2uoU7VOt1AAITsPRPH9n",
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "code": "invalid_value",
    "message": "Invalid value: 'qwen-tts'. Supported values are: 'Qwen2.5-tts'.",
    "param": "session.input_audio_transcription.model",
    "event_id": "event_123"
  }
}

session.created

The server returns this event immediately after a client connects. This event contains the default server configurations for the connection.

Parameter	Type	Description
type	string	The event type. The value is always `session.created`.
session	object	The session configuration.
session.id	string	The unique ID of the session.
session.object	string	The service name of the session.
session.mode	string	The response mode of the model.
session.model	string	The model used.
session.voice	string	The voice used to generate the audio.
session.response_format	string	The audio output format. Currently, only pcm is supported.
session.sample_rate	integer	The audio output sample rate. Currently, only 24000 is supported.

{
  "event_id": "event_xxx",
  "type": "session.created",
  "session": {
    "object": "realtime.session",
    "mode": "server_commit",
    "model": "qwen-tts-realtime",
    "voice": "Cherry",
    "response_format": "pcm",
    "sample_rate": 24000,
    "id": "sess_xxx"
  }
}

session.updated

The server returns this event after it receives and successfully processes a session.update request from the client. If an error occurs, an error event is returned instead.

Parameter	Type	Description
type	string	The event type. The value is set to `session.updated`.
session	object	The session configuration.
session.id	string	The unique ID of the session.
session.object	string	The service name of the session.
session.mode	string	The output mode of the model.
session.model	string	The model used.
session.voice	string	The voice of the generated audio.
session.language_type	string	Specifies the language for speech synthesis. The default value is `Auto`. `Auto`: Use this value if the language of the text is uncertain or contains multiple languages. The model automatically detects and pronounces text segments in different languages, but the pronunciation may not be perfectly accurate. Specific language: Use this option if the text is in a single language. Specifying a language significantly improves the synthesis quality and typically yields better results than `Auto`. Valid values include the following: `Chinese` `English` `German` `Italian` `Portuguese` `Spanish` `Japanese` `Korean` `French` `Russian`
session.response_format	string	The format of the audio output from the model. Currently, only pcm is supported.
session.sample_rate	integer	The sample rate of the audio output from the model. Currently, only 24000 is supported.

{
  "event_id": "event_xxx",
  "type": "session.updated",
  "session": {
    "id": "sess_xxx",
    "object": "realtime.session",
    "model": "qwen-tts-realtime",
    "voice": "Cherry",
    "language_type": "Chinese",
    "mode": "commit",
    "response_format": "pcm",
    "sample_rate": 24000
  }
}

input_text_buffer.committed

The server returns this event after receiving an input_text_buffer.commit event from the client.

Parameter	Type	Description
event_id	string	The ID of the event.
type	string	The type of the event. The value is fixed to `input_text_buffer.committed`.
item_id	string	The ID of the user message item to be created.

{
  "event_id": "event_FC6MA88wS2oEeXkPvWsxX",
  "type": "input_text_buffer.committed",
  "item_id": ""
}

input_text_buffer.cleared

This response event is sent by the server after the client sends the input_audio_buffer.clear event.

Parameter	Type	Description
event_id	string	The ID of the event.
type	string	The event type. The value is fixed as `input_text_buffer.cleared`.

{
    "event_id": "event_1122",
    "type": "input_text_buffer.cleared"
}

response.created

The server returns this event when it generates a new model response.

Parameter	Type	Description
type	string	Set to `response.created`.
event_id	string	The ID of the event.
response	object	The response object.
response.id	string	The unique ID of the response.
response.object	string	The object type. Set to `realtime.response`.
response.status	string	The final status of the response. Valid values are completed, failed, in_progress, and incomplete.
response.voice	string	The voice used for the generated audio.
response.output	array	This parameter is currently empty.

{
  "event_id": "event_IMnLqDvG6Ahhk7sWV2uOs",
  "type": "response.created",
  "response": {
    "id": "resp_USvBwHktHcz76r6GaIJUV",
    "object": "realtime.response",
    "conversation_id": "",
    "status": "in_progress",
    "voice": "Cherry",
    "output": []
  }
}

response.output_item.added

The server returns this event when a new item is added to the output.

Parameter	Type	Description
type	string	The value is fixed at `response.output_item.added`.
response_id	string	The ID that associates all outputs from the same response.
output_index	integer	The index of the output item in the response. The value is currently fixed at 0.
item	object	Information about the output item.
item.id	string	The unique ID of the output item.
item.object	string	The value is fixed at `realtime.item`.
item.status	string	The status of the output item.
item.content	array	The content of the message.

{
  "event_id": "event_INDGnGNulaXCrStd9ZM5X",
  "type": "response.output_item.added",
  "response_id": "resp_USvBwHktHcz76r6GaIJUV",
  "output_index": 0,
  "item": {
    "id": "item_FIrYGaNVK3rbIZqeY4QjM",
    "object": "realtime.item",
    "type": "message",
    "status": "in_progress",
    "role": "assistant",
    "content": []
  }
}

response.content_part.added

The server returns this event when a new content part is added to the output.

Parameter	Type	Description
type	string	The value is fixed as `response.content_part.added`.
response_id	string	The ID of the response.
item_id	string	The ID of the message item.
output_index	integer	The index of the output item in the response. The value is currently fixed at 0.
content_index	integer	The index of the content part within the response output item. The value is currently fixed at 0.
part	object	The completed content part.
part.type	string	The type of the content part.
part.text	string	The text of the content part.

{
  "event_id": "event_DigZ95MWN36YYyyjcENoq",
  "type": "response.content_part.added",
  "response_id": "resp_USvBwHktHcz76r6GaIJUV",
  "item_id": "item_FIrYGaNVK3rbIZqeY4QjM",
  "output_index": 0,
  "content_index": 0,
  "part": {
    "type": "audio",
    "text": ""
  }
}

response.audio.delta

The server returns this event when the model incrementally generates new audio data.

Parameter	Type	Description
type	string	The value is fixed at `response.audio.delta`.
response_id	string	You can associate all outputs from the same response.
item_id	string	The ID of the message item. This ID is used to associate all parts of the same message item.
output_index	integer	The index of the output item in the response. The value is currently fixed at 0.
content_index	integer	The index of the internal part within the response output item. The value is currently fixed at 0.
delta	string	The incremental audio data that is output by the model. The data is Base64-encoded.

{
  "event_id": "event_B1osWMZBtrEQbiIwW0qHQ",
  "type": "response.audio.delta",
  "response_id": "resp_B1osWTzBb8hO0WsELHgVP",
  "item_id": "item_B1osWH81fXDoyim1T5fsF",
  "output_index": 0,
  "content_index": 0,
  "delta": "base64 audio"
}

response.content_part.done

The server returns this event when the output for a content part is complete.

Parameter	Type	Description
type	string	The value is fixed at `response.content_part.done`.
response_id	string	The ID of the response.
item_id	string	The ID of the message item.
output_index	integer	The index of the output item in the response. The value is fixed at 0.
content_index	integer	The index of the content part within the response output item. The value is fixed at 0.
part	object	The completed content part.
part.type	string	The type of the content part.
part.text	string	The text of the content part.

{
  "event_id": "event_Vo2YUjlYQJ4colH8nVzkU",
  "type": "response.content_part.done",
  "response_id": "resp_USvBwHktHcz76r6GaIJUV",
  "item_id": "item_FIrYGaNVK3rbIZqeY4QjM",
  "output_index": 0,
  "content_index": 0,
  "part": {
    "type": "audio",
    "text": ""
  }
}

response.output_item.done

The server returns this event when the output for an item is complete.

Parameter	Type	Description
type	string	The value is always `response.output_item.done`.
response_id	string	The ID of the response.
output_index	integer	The index of the output item in the response. The value is 0.
item	object	Information about the output item.
item.id	string	The unique ID of the output item.
item.object	string	The value is always `realtime.item`.
item.status	string	The status of the output item.
item.content	array	The content of the message.

{
  "event_id": "event_LO6SJRKIQ9NBayyYB8a1A",
  "type": "response.output_item.done",
  "response_id": "resp_USvBwHktHcz76r6GaIJUV",
  "output_index": 0,
  "item": {
    "id": "item_FIrYGaNVK3rbIZqeY4QjM",
    "object": "realtime.item",
    "type": "message",
    "status": "completed",
    "role": "assistant",
    "content": [
      {
        "type": "audio",
        "text": ""
      }
    ]
  }
}

response.audio.done

The server returns this event when the model finishes generating audio data.

Parameter	Type	Description
type	string	The value is `response.audio.done`.
response_id	string	The ID of the response. This ID associates all outputs that belong to the same response.
item_id	string	The ID of the message item. This ID associates all parts of the same message item.
output_index	integer	The index of the output item in the response. The value is 0.
content_index	integer	The index of the content part within a response output item. The value is 0.

{
  "event_id": "event_LZaOHPzXYMUXGBcVkBmKX",
  "type": "response.audio.done",
  "response_id": "resp_USvBwHktHcz76r6GaIJUV",
  "item_id": "item_FIrYGaNVK3rbIZqeY4QjM",
  "output_index": 0,
  "content_index": 0
}

response.done

The server returns this event when the response generation is complete. The Response object in this event contains all output items, but does not include the raw audio data that has already been returned.

Parameter	Type	Description
type	string	The value is fixed at `response.done`.
response	object	The response object.
response.id	string	The unique ID of the response.
response.object	string	The object type. The value is fixed at `realtime.response`.
response.output	array	The output of the response.
response.usage	object	The billing details for the speech synthesis.
response.usage.characters	integer	The number of billable characters for Qwen3-TTS Realtime.
response.usage.total_tokens	integer	The total number of tokens for the Qwen-TTS Realtime input and output (synthesized audio).
response.usage.input_tokens	integer	The total number of tokens for the Qwen-TTS Realtime input.
response.usage.output_tokens	integer	The total number of tokens for the Qwen-TTS Realtime output.
response.usage.input_tokens_details	integer	Details of the tokens in the Qwen-TTS Realtime input.
response.usage.input_tokens_details.text_tokens	integer	The total number of tokens for the Qwen-TTS Realtime text input.
response.usage.output_tokens_details	integer	Details of the tokens in the Qwen-TTS Realtime output.
response.usage.output_tokens_details.text_tokens	integer	The total number of tokens for the Qwen-TTS Realtime text output.
response.usage.output_tokens_details.audio_tokens	integer	The total number of tokens for the Qwen-TTS Realtime audio output. The audio is converted to tokens at a rate of 50 tokens per second. Audio with a duration of less than one second is counted as 50 tokens.

Qwen3-TTS Realtime

{
    "event_id": "event_Aemy83XqHFFDDSeJIDn6N",
    "type": "response.done",
    "response": {
        "id": "resp_LFeR42yXZ9SxUAeXjmyTz",
        "object": "realtime.response",
        "conversation_id": "",
        "status": "completed",
        "modalities": [
            "text",
            "audio"
        ],
        "voice": "Cherry",
        "output": [
            {
                "id": "item_Ae1lv2XmRljRSG96L8Zm1",
                "object": "realtime.item",
                "type": "message",
                "status": "completed",
                "role": "assistant",
                "content": [
                    {
                        "type": "audio",
                        "transcript": ""
                    }
                ]
            }
        ],
        "usage": {
            "characters": 25
        }
    }
}

Qwen-TTS Realtime

{
  "event_id": "event_xxx",
  "type": "response.done",
  "response": {
    "id": "resp_xxx",
    "object": "realtime.response",
    "conversation_id": "",
    "status": "completed",
    "modalities": [
      "text",
      "audio"
    ],
    "voice": "Cherry",
    "output": [
      {
        "id": "item_FIrYGaNVK3rbIZqeY4QjM",
        "object": "realtime.item",
        "type": "message",
        "status": "completed",
        "role": "assistant",
        "content": [
          {
            "type": "audio",
            "transcript": ""
          }
        ]
      }
    ],
    "usage": {
      "total_tokens": 67,
      "input_tokens": 3,
      "output_tokens": 64,
      "input_tokens_details": {
        "text_tokens": 3
      },
      "output_tokens_details": {
        "text_tokens": 0,
        "audio_tokens": 64
      }
    }
  }
}

session.finished

The server returns this event when the generation of all responses is complete.

Parameter	Type	Description
type	string	The value is fixed at `session.finished`.
event_id	object	The ID of the event.

{
  "event_id": "event_2239",
  "type": "session.finished"
}