Timed Text Speech

Instantly create highly accurate captions and subtitles
with Machine Learning based speech to text technology.


Why use Timed Text Speech

Get results in minutes instead of days, save time by reviewing results online while the speech engine continues processing
Easy to use
Just upload a media file, we transcribe the audio and return timed text in a variety of formats (SRT, SCC)
Cost effective
Priced per minute of transcribed media, pay only for what you use
Supports multiple languages including Arabic, Brazilian Portuguese, English (UK and US), French, Japanese, Chinese, Spanish (Castilian, Latin American, and North American)
Highly accurate transcription
Improve results using custom vocabularies (names, acronyms, places, specialized terms)
Integrates with existing tools
Transcribe directly from within MacCaption or CaptionMaker
Who should use Timed Text Speech?

Content creators in need to produce captions and subtitles for time-sensitive content that must be aired on TV or the Internet with short turnaround times, as well as short-form content such as promos that require captioning.

Captioning service companies looking for time-efficient solution to speed up their service with cost-effective and easy-to-use automated transcription option.

Corporations, government organizations and educational institutions wishing to bring captioning in-house.

Sign up
How it works

Timed Text Speech uses the latest Machine Learning (ML) based technology to accurately transcribe human speech (extracted from a video) into time stamped text.

The transcript can then be edited and formatted for a variety of caption and subtitle standards.

Using the Telestream Cloud console:
  • Create a transcription project, add optional vocabulary terms and select a timed text file format (SRT, JSON, CSV)
  • Upload a video file to the project
  • Telestream Cloud extracts the audio and the transcribed results appear in the web console
  • Review and edit the transcription
  • Download the timed text file