Uploading audio data via the SpeechSense interface
This guide will help you upload audio data to SpeechSense via its interface. Use this option to quickly upload data and get started with SpeechSense. Use the API to fully upload your data.
If you want to upload the chat text instead of voice call audio, follow this guide.
Getting started
-
Create a connection of the Two-channel audio type.
-
Create a project with the new connection.
Voice call recordings will be uploaded to the project and connection you created.
Uploading data
-
Open the SpeechSense home page
. -
Go to the space you need and select a project.
-
In the top-right corner, click Adding dialogs.
-
In the window that opens, select files to upload by doing one of the following:
- Select one or more files in the file manager and drag them to the SpeechSense area marked by the dotted line.
- Click Select files and select one or more files.
Note
Via the interface, you can only upload MP3 audio files. Use the API to upload audio files in other formats supported by SpeechSense.
Once you select the files, their upload into the intermediate buffer will begin. Wait for this operation to complete.
-
Fill in the metadata for each file:
-
Start date: Specify the dialog start date.
-
Direction:
- Select Outgoing if the dialog was initiated by the agent.
- Select Incoming if the dialog was initiated by the customer.
-
Language: Specify language.
-
Agent: Enter the agent's name.
-
operator_id: The agent ID is generated automatically and can be corrected as needed.
-
Client: Enter the customer name.
-
client_id: The customer ID is generated automatically and can be corrected as needed.
-
-
Click Add \ <N> dialogs. This will start uploading dialogs into the project followed by speech recognition and analysis. This may take some time.