Don't forget to enter your Broadcast Asia offer code for 10% off the 1 Year Subscription option*
- Frame-accurate digital media playback with frame-step audio scrubbing, audio waveform display and speech presence detection
- Import/export of common file formats including PAC, STL, 890, SRT and Timed Text XML,
- Unicode language support and multilingual spell-checking
- Advanced timing rules including automatic adjustment for shot changes
- Automatic text-to-speech alignment (PRO version, which also features the option to use speech recognition for faster text entry).
This revolutionary tool is capable of parsing almost any script layout to extract dialogue text and speaker cues given just a few minutes’ worth of guidance.
A major time-saver for for same language subtitling, this text-to-speech alignment function (WINCAPS Q4 Pro) allocates timecodes to subtitles based on equivalent programme speech timing and taking account of timing rules for shot changes, reading speed, inter-subtitle gaps, durations and subtitle sequences. This powerful facility truly can deliver a quantum leap in subtitling productivity. Currently available for Dutch, English, Finnish, French, German, Italian, Spanish and Swedish (please ask about support for other languages).
Re-speaking can be used to transcribe a media file, giving a significant speed improvement compared to retyping. WINCAPS Q4 Pro can also support re speaking with automatic alignment for maximum benefit. The VOCAB FINDER utility allows speech recognition users to research chosen parts of the internet for new vocabulary that may be needed in subtitles. This can be particularly helpful for finding proper nouns, such as place names and names of people, and acronyms related to a specific event or news story, for example.
In an enterprise setting, the DataStore reference database significantly assists workflow by gathering and making available to a subtitler all the relevant information needed for a given job, for example the customer template, programme word lists, abbreviation files, language rules for text formatting, path to media file and path to scripts.
Fed by your ingest system, the Indexer server application automatically processes each new media file to create helper files that identify shot changes and audio information, ready for loading by the subtitling client.
Rather than tying up a client workstation, DVD Builder is a back-end application that takes completed subtitle files from WINCAPS Q4 clients and queues them up for processing into DVD asset and control files.
WINCAPS can be used to produce subtitles easily, efficiently and accurately for any recorded media. It supports a comprehensive range of media and file formats, ensuring compatibility with all production and delivery requirements. WINCAPS blends the human skills of the subtitler or translator with state-of-the art audiovisual analysis and speech processing. This delivers high productivity and greater accuracy – the advanced features assist in capturing accurate timecodes, checking spelling and timing, and ensuring that files are consistent with client specifications.
One of the advanced productivity features of WINCAPS Q4 Pro is AutoTime, which aligns subtitle text against the programme audio track in supported languages. This can greatly speed up the task of timing subtitles provided the dialogue is reasonably free from background noise.
AutoTime is a key feature of the new WINCAPS Q4 enterprise productivity suite for offline subtitling. Along with a brand new user interface, WINCAPS Q4 provides better support for faster workflows. For instance, an imported script can be easily aligned and automatically coloured; and respoken speech-input can be used for timed transcription to create a timed first-draft offline subtitle file after just one re-speaking pass through the programme.
Speech recognition with a respeaker for live subtitling
Speaktitle Speech Recognition for WINCAPS Live Subtitle Preparation High quality, real-time speech recognition tools, such as Dragon NaturallySpeaking®, now provide a viable and economic alternative to conventional keyboard-based methods for live subtitling. Indeed it has become the tool of choice for many because of the reduced costs and training times and Screen has been at the forefront of the introduction of this technology for television subtitling.
The use of speech recognition for subtitling does not mean recognising the voice of the TV presenter – the technology is not yet good enough to deliver sufficient accuracy. Instead, a trained “re-speaker” listens to the programme, and re-dictates a slightly edited transcript of what is said into a speech recognition system that has been trained for their voice. The re-speaker is trained to achieve high accuracy and consistency.
Integrated Speech Recognition
WINCAPS supports speech recognition for both live and offline transcription with an integrated speech input interface that is optimised for Dragon NaturallySpeaking® , the most popular speech recognition system (other real-time speech recognition engines may also be supported using a standard keyboard emulation interface). WINCAPS adds further value by enforcing a house style, interpreting spoken style control commands, smoothing the delivery rate and providing for keyboard intervention/correction if required.
- Maintains text input path regardless of application focus
- Corrects consistently misrecognised words or phrases
- Improved delivery time
- Reduced loading on system resource
- Keyboard intervention without pause in re-speaking
- Dragon NaturallySpeaking® is a registered trademark of Nuance, Inc. and is used here under license.
Using speech recognition for offline workflows too
As well as its use for live subtitling in our WINCAPS Q-LIVE software, speech recognition is used for offline preparation too. Screen’s WINCAPS Q4 has the same speech input handling as its live subtitling counterpart with AutoTime to enable automatic timing alignment of the subtitles against the audio track, allowing a respeaker to create a viable, timed first-draft offline subtitle file after just one pass through the programme.
Automatic subtitling using speech recognition
Speech technology is still a long way from converting soundtrack audio into accurate text in real-time. The challenges are multiple and include: sound effects and background music,different voices and accents and switching between them, unexpected vocabulary and punctuation. Even with clean speech audio from a single speaker and slower than real-time processing the results are still rather unpredictable – to the point where our experiments indicate that it’s quicker to transcribe using a re-speaker rather than trying to correct the automatic transcription from the soundtrack.
Using an automatic transcription to timecode a script
Screen has done some research on using automatic speech recognition to enable the production of well-timed subtitles from an untimed script; the former containing less than perfect text that’s nevertheless reasonably accurately time-stamped, while the latter contains accurate text and punctuation but no timing. The results are very promising – ask us for more information.
WINCAPS exports for digital workflows
The standard WINCAPS off-line subtitling software has several export formats for digital workflows. These include: DCI and DTS digital cinema, Final Cut Pro, DVD Studio Pro, SMI web subtitles, and the W3C Timed Text format (DFXP/MXF).
DVD Builder for DVD and Blu-ray disc
The DVD Builder (an add-on software module for WINCAPS) creates DVD subtitle assets as a sequence of image files and associated control files for all popular DVD authoring systems. It includes user-configurable templates, and a batch processing option for unattended workflow.
MediaMate – encoding subtitles into media files
Adding subtitle data directly into a media file as part of an automated file-based workflow saves time, effort and avoids expensive broadcast hardware. MediaMate supports a wide variety of subtitle/caption types, media file formats and workflows. More on MediaMate.
CAPture – restore subtitles from tape or capture live output
CAPture enables teletext subtitle data to be captured from VBI (e.g. off-air or tape) and compiled into a subtitle file for archiving, repeat broadcast or amendment. Both scrolling (word-by-word) and block mode subtitles can be handled, with a conversion from scrolling format into block for subsequent re-editing and timing. This has great time-saving benefits in two situations: (1) restoring editorial access to VBI-encoded subtitle data in video archives, and (2) enabling easy repeat use of subtitles from a live broadcast, even if some editing is needed.