tencent cloud

All product documents
Video on Demand
DocumentationVideo on DemandProduct IntroductionFeaturesMedia IntelligenceSmart Subtitles
Smart Subtitles
Download PDF
Last updated: 2024-11-04 10:46:28
Smart Subtitles
Last updated: 2024-11-04 10:46:28
Download PDF

Overview

With VOD II capabilities, you can automatically generate multilingual subtitle files for video content. This is suitable for short dramas aimed at international markets, multilingual online education courses, and organizing multinational audio and video conference minutes.
The smart subtitling feature includes the Full Speech Recognition and Voice Translation capabilities. The former uses advanced ASR algorithms combined with massive training data, providing high accuracy even in conditions with ambient noise and background sounds. The latter leverages the advantages of Neural Machine Translation (NMT) and Statistical Machine Translation (SMT), learning translation knowledge from large multilingual corpora to achieve high-accuracy translations from the source language to target languages. Currently, it supports translation for dozens of languages, including Chinese, English, Japanese, Korean, French, Spanish, Italian, German, Turkish, Russian, Portuguese, Vietnamese, Indonesian, Thai, Malay, Arabic, and Hindi.

Suitable Scenarios

Voice Translation Scenarios

Scenario
Description
Short Dramas Going Global
For short dramas targeting international audiences, smart subtitling supports efficient output of multilingual translated subtitles, increasing the efficiency of generating global versions of short dramas.
Movie/TV series
When movies and TV series are released in different countries or regions, adding local language subtitles helps in their dissemination. Smart Translation Subtitles can automatically generate subtitles in the target country's or region's language, which greatly improves efficiency compared to manual subtitling.
Multinational Audio/Video Meetings
For key multinational meetings, recordings can have participants' native language subtitles automatically generated to help with organizing meeting minutes and viewing.
Multilingual Recorded Courses Learning
When online education courses are released in different countries or regions, smart translation subtitles can automatically generate local language subtitles, making it easier to understand and learn.

Speech Full Text Recognition Scenarios

Scenario
Description
Online classes
Smart subtitling can automatically generate subtitles for online course recordings to make it easier for viewers to learn the content of the course.
Social media platforms
General users can add subtitles to their vlogs and share them with others.
Movie/TV series
Old videos are often not subtitled, which makes it difficult for some viewers to enjoy them. The smart subtitling feature can automatically generate subtitles for such videos, which is much more efficient than manual subtitling.
Audio/Video meetings
Subtitles can be automatically generated for recordings of important meetings to facilitate minutes generation. Subtitles also make it easier for viewers to navigate meeting recordings.
Speech quality control
Subtitles can be generated for call recordings to help call centers quickly evaluate the quality of customer service, improving the efficiency of speech quality assurance.

How to Use

Method 1: Launching tasks via the console

You can launch Speech Recognition tasks via the VOD console. Use templates to specify the recognition items as either "Speech Full Text Recognition" or "Voice Translation" for the tasks. Once the task is complete, refresh the subtitle list to view the recognized or translated subtitle files. For detailed instructions, please refer to How to Add Subtitles to Videos.

Method 2: Launching tasks via API

Launching Voice Translation Tasks

Intelligent subtitles are implemented through the audio and video CR feature of speech translation. Specific usage steps:
1. Create an audio and video CR template, enable the speech translation configuration option (AsrTranslateConfigure) switch (Switch), set the media source language (SrcLanguage) and the translation target language (DstLanguage), specify the subtitle file format (SubtitleFormats) and set the subtitle name (SubtitleName):
{
 "AsrTranslateConfigure": {
     "Switch": "ON",
     "SrcLanguage": "zh",
     "DstLanguage": "en",
     "SubtitleFormats": ["vtt", "srt"],
     "SubtitleName": "test"
 }
}
Get the template ID from the response.
2. With the template ID obtained in step 1, initiate a smart subtitling task as instructed in Task Initiation.
3. Get the task result as instructed in Getting Result.

Initiate a full speech recognition task

You can implement subtitles through the Video Content Recognition full speech recognition feature as follows:
1. Create an audio/video content recognition template, enable the full speech recognition configuration item (AsrFullTextConfigure) switch (Switch), set the media source language (SrcLanguage), specify the format for the generated subtitle file (SubtitleFormats) and set the subtitle name (SubtitleName):
{
 "AsrFullTextConfigure": {
     "Switch": "ON",
     "SrcLanguage": "zh",
     "SubtitleFormats": ["vtt", "srt"],
     "SubtitleName": "test"
 }
}
Get the template ID from the response.
2. With the template ID obtained in step 1, initiate a smart subtitling task as instructed in Initiating a Task.
3. Get the task result as instructed in Getting the Result.
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 available.

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