South Korea's Naver Cloud, the cloud computing arm of Naver Corp., announced on Friday that it has launched a "real-time streaming" feature for "CLOVA Speech," an enterprise-level speech-to-text service that converts voice into text.
The AI-based CLOVA Speech service is currently applied in automatic subtitle generation for Naver News, Naver Shopping Live and customer service call data operations.
The new feature extracts and generates subtitles immediately from the speaker's voice in real-time streaming videos, supporting Korean, English and Japanese.
By utilizing real-time streaming technology, it is possible to broadcast live captions without the need for separate typing tasks for stenographers. In customer service centers, monitoring the immediate text transcription of phone conversations can lead to faster customer responses.
In conjunction with the real-time streaming launch, Naver Cloud has reduced the cost of "voice recognition & speaker recognition" by 40% and segmented fees based on functions such as voice recognition, speaker recognition and event detection, eliminating the previous flat-rate pricing. An optional feature evaluating the accuracy of English pronunciation has also been added.