TVEyes Language Technology Labs
TVEyes is the only global audio and video intelligence company with its own proprietary speech-to-text software and technology lab.
What We Do
Through an innovative and collaborative partnership, PerVoice and TVEyes founded TVEyes Language Technology (TLT) Labs.
There, teams of developers, linguists and technologists created our proprietary speech-to-text (STT) software. This innovative software enables us to transcribe more than 2 million hours of content monthly from our global broadcast capture network and podcast sources.
At TLT labs, we also develop, test and train language models, which allow us to transcribe audio and video streams and produce transcripts in different languages for end users. This uniquely enables TVEyes to rapidly launch media monitoring and data solutions in new countries and languages to meet customer requirements.
Today, we have developed nearly 40 different language models, which we use every day for customers, and even more are in development.
Why it Matters
TLT Labs is one more way we ensure that our customers can “find it here” with TVEyes – no matter what “it” sounds like to the ear.
TLT Labs continuously updates its language models to improve speech-to-text accuracy, transcription models, punctuation and more.
TLT Labs currently offers nearly 40 language models with more in development. In addition to English, we offer Mandarin, Standard Arabic, Japanese, French and Spanish. TLT Labs enables TVEyes to add new languages to our platform with unprecedented speed and accuracy and launch new markets with rapid turnaround.
Our partners are empowered to submit keywords for improved recognition and continued STT training to ensure quality transcriptions and translations.