Sambert server-sent events
User guide: For model introduction and selection recommendations, see Speech synthesis.
ImportantAlibaba Cloud Model Studio has released a workspace-specific domain for the China (Beijing) region: https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com. The new dedicated domain delivers superior performance and higher stability for inference requests. We recommend migrating from https://dashscope.aliyuncs.com to the new domain.
{WorkspaceId} is your workspace ID, which can be found on the Workspace Details page in the Alibaba Cloud Model Studio console. The existing domain remains fully functional.
task-started
When the client sends the run-task command, the server returns a task-started event to indicate the task has started.
header.task_id The task ID generated by the client. header.event The event type. Fixed value: payload Empty object with no content. | |
result-generated
After the client sends text, the server streams result-generated events. Each event includes full timestamp data in output.sentence, with word-level and phoneme-level timestamps when the corresponding parameters are enabled.
header.task_id The task ID generated by the client. header.event The event type. Fixed value: payload.output The synthesized audio output. payload.usage Billing usage for the current event. | |
task-finished
When the task is complete, the server returns a task-finished event. The client can then close the WebSocket connection or reuse the connection for a new task.
header.task_id The task ID generated by the client. header.event The event type. Fixed value: payload.usage.characters The cumulative billed character count up to the current event. | |
task-failed
When a task fails, the server returns a task-failed event. Close the WebSocket connection and handle the error.
header.task_id The task ID generated by the client. header.event The event type. Fixed value: header.error_code The error code. header.error_message The detailed error message. | |