HOW is VIDEOSPACE SEARCH Engine different other search engines?
In short, besides the video title and metadata, the VideoSpace Search Engine searches INSIDE the video for its spoken content.
How does VideoSpace search our videos?
We will Machine Transcribe your video and video index the results. Then we put the indexes into our Video.Search Engine to perform keyword search.
We are currently looking to add Video OCR to our indexing capabilities, which will enhance the discoverability of your video content.
HOW ACCURATE IS OUR Video.SEARCH engine?
Under ideal conditions, accuracy can go as high as 95%. However, there are a range of factors (like accent, speech volume, diction, background noise, multiple voices speaking at the same time, etc) that can affect the accuracy rate.
HOW MANY LANGUAGES DOES vIDEO.sEARCH SUPPORT?
We are to machine-transcribe speech-to-text and search in the following languages:
- British English
- Mexican English
- Portuguese (Brazilian)
- Arabic (Egyptian)
However, if we enable our Video.Translation process, we can search the videos in more than 50 languages!
CAN WE IMPROVE THE TRANSCRIBER'S Speech-to-text ACCURACY?
Yes! You can. Unlike other machine transcription engines, you can give us a set of keywords that is associated with your audio/video file to be transcribed. For example, scientific and medical terms, a person's name or even a technical part name, etc. This is what we do with these terms:
- Our Engine will rank these words/terms into a higher order and use them first when a similar sounding word appears.
- We use these words as a search term to search the web for related documents and use these documents to expand its our Transcription Engine's vocabulary with the final aim to improve search results.
We know of no other Machine Transcription Service that allows you to improve the accuracy!
ARE THERE AUDIO FILES THAT IS "UNSUITABLE"?
We do not want you to waste your money and our effort in doing a job that will yield bad results. Ideally, your media files should have very clear speech (without background music, noise, effects, or microphone hiss). Examples of appropriate content are: recorded meetings, lectures or presentations.
The following might not be suitable for transcribing: movies, TV shows, anything with mixed audio and sound effects, poorly recorded content with background noise (hiss). In addition, there are factors that can affect the accuracy. For example, factors like:
- heavy localised accent
- low speech volume
- bad diction
- heavy background noise
- multiple voices speaking at the same time
For example, if the accent is really heavy, the result can be as low as 10% accuracy. That will be a total waste of resources. If in doubt, please use a human transcription service instead.
DO WE PROVIDE HUMAN TRANSCRIPTION SERVICES?
No. We provide only machine transcription.
HOW MUCH DOES IT COST?
For pricing details, please CONTACT US.
WHAT FILE FORMAT DO YOU ACCEPT?
We accept the following file formats: .wav, .wma, .wmv, .mp3, .mp4, .m4a, .aac. We do not accept files on physical devices such as SD cards, CDs, thumbdrives or tapes. If you have a different file format than what is listed here, please contact us
IS MY DATA SECURED?
Yes! Your privacy is of utmost importance! Files are stored using industrial grade encryption. Absolutely nobody outside VideoSpace has access to our systems and your files. Even within VideoSpace, we do NOT look into your files unless it is absolutely necessary and with your permission.
Integrity and trust is the cornerstone of VideoSpace.