Speech transcription

更新时间:
复制 MD 格式

What is speech transcription

You can transcribe audio files from a dataset and export the resulting conversation text. This is a paid service. You will be charged again if you re-transcribe audio files.

Feature access

Log in to the Smart Conversation Analysis console. In the navigation pane, choose Smart Tools > Speech Transcription.

Create a speech transcription task

Click Create Transcription Task and configure the task parameters.

  • Task Name: Enter a custom name for the task. This parameter is required.

  • Dataset: Select an uploaded dataset to use as the source data. This parameter is required. For more information, see Dataset management.

  • Speech Model: Select an existing speech model. This parameter is required. For more information, see Language models.

  • Hotword Model: Select an existing hotword model. This parameter is optional. For more information, see Hotword models.

  • Execution Mode: Select Execute Immediately or Scheduled Execution. This parameter is required.

    • Scheduled Execution: If you select this option, you must set an execution time. The time must be in the future.

  • Priority: Set the priority to Low, Medium, or High.

Speech transcription task list

The list contains the following fields:

Field Name

Description

Task Name

The name of the task. You specify this name when you create the task.

Dataset

The dataset used for the task. You select this dataset when you create the task.

File Count

The number of audio files in the dataset for this task.

Creation Time

The time when the task was created.

Priority

The execution priority of the task.

Execution Status

The execution status of the task. Blue indicates that the task is pending. Green indicates that the task is complete. Red indicates that an error occurred.