tencent cloud

All product documents
StreamLive
DocumentationStreamLive Feature GuideCaptionSmart Subtitling
Smart Subtitling
Download PDF
Last updated: 2024-07-22 10:38:01
Smart Subtitling
Last updated: 2024-07-22 10:38:01
Download PDF
The Smart Subtitling function converts voice information into subtitles by real-time Automatic Speech Recognition (ASR) during live streaming, and subsequently translates them into the target language. Currently, this feature supports four languages: Chinese, English, Japanese, and Korean. In practical applications, please select the appropriate language according to your business needs and audience.

Points of Attention

Currently, the Smart Subtitling feature can only be used in joint transcoding.
The ability to generate the intelligent captions is provided by Media Processing Service (MPS) to StreamLive. When using it for the first time, you need to authorize MPS to access StreamLive's data to generate captions.
When using the intelligent caption feature in StreamLive, in addition to StreamLive's live transcoding fees, there will also be speech recognition fees from MPS. Translating across languages will incur speech translation fees from MPS. For specific billing information, please refer to the MPS Billing Document.

Prerequisites for Use

You have activated the StreamLive service.

Configuring Smart Subtitles

1. Log in to StreamLive Console, navigate to Channel Configuration and Configure Output Groups. For the Output you want to configure, click Setting to enter Transcoding Settings.



2. Select Joint Transcoding, and click Add Caption.



3. Scroll to the bottom of the page, and in the Caption module, choose the caption source as Analysis.



Note：
Since the feature is provided by MPS to StreamLive, role authorization is required to support MPS in obtaining StreamLive data and generating captions.
When in use, the system will verify if you have authorized it before. If you have, you can directly configure captions.
If you have not authorized before, the system will guide you through role authorization. Once you agree, you can continue configuring captions.
4. Configure caption



Configuration Item
Description
Format
Currently, only burn in is supported, which directly merges subtitles into the video.
Source language
It supports recognition of four source languages: Chinese, English, Japanese, and Korean.
Content type
Source: display only source language.
Source + Target: display source language and translation language.
Target: display only translation language.
Target Language
Currently, the source language can be translated into three target languages.
Dynamic/Steady State Effect
Default setting: Delayed Steady State Subtitles. The system will delay the live streaming according to the set time, but the experience of watching complete sentence is better. The default delay time is 10 seconds, with other options including 20 seconds, 30 seconds, and 60 seconds.
You can also choose: Real-time Dynamic Subtitles, which has a shorter delay, but the captions will dynamically correct the content word-by-word according to the spoken content.
Line Spacing
The vertical distance percentage represents the ratio of the vertical distance to the vertical width of the subtitles from the bottom of the screen. The vertical position of the subtitles is configured by adjusting the vertical distance percentage. Value range: 0 - 100%.
Margins
The horizontal distance percentage represents the ratio of the horizontal distance to the horizontal width of the subtitles and the side of the screen. The horizontal position of the subtitles is configured by adjusting the horizontal distance percentage. Value range: 0 - 50%.
Lines
Options are 1 or 2, with 1 selected by default.
When lines exceed the displayed range, only the latest content will be displayed.
Characters Per Line
Value range: 1-100. Default value: 45. One Chinese character count as one word. One English character or number counts as half a word. The fewer the characters per line are, the larger the font size is.
Font
Options include Heiti, SimSun, Dynacw Diamond Black, and Helvetica.
The font selection will vary depending on the Source language and Target language.
Color
The font color is white by default. The color can be customized.
The background color is black by default. The customization of level of transparency is supported.
5. Preview
Turn on the preview switch, enter test text, and the preview effect will be displayed according to your previous configuration. Additionally, you can readjust the resolution of the preview screen.



6. Click Confirm to save your current Transcoding Configuration and Caption Configuration.





Was this page helpful?
You can also Contact Sales or Submit a Ticket for help.
Yes
No

Feedback

Contact Us

Contact our sales team or business advisors to help your business.

Contact Us
Technical Support

Open a ticket if you're looking for further assistance. Our Ticket is 7x24 avaliable.

Submit a Ticket
7x24 Phone Support
Hong Kong, China
+852 800 906 020 (Toll Free)
United States
+1 844 606 0804 (Toll Free)
United Kingdom
+44 808 196 4551 (Toll Free)
Canada
+1 888 605 7930 (Toll Free)
Australia
+61 1300 986 386 (Toll Free)
EdgeOne hotline
+852 300 80699
More local hotlines coming soon