What video formats does Fluen Studio support?
Fluen Studio supports essentially any playable video file, including common formats like .mp4, .mov, .m4v, .mkv, and .avi. For audio files, Fluen Studio is compatible with formats such as .mp3, .wav and .aac.
Which language pairs do you support?
Fluen Studio offers support for 33 different languages at this time. We can handle captioning and translation both from and into the following languages: English, Bulgarian, Chinese, Chinese (Taiwan), Czech, Danish, Dutch, Estonian, Finnish, French, French (Canada), German, Greek, Hungarian, Indonesian, Italian, Japanese, Korean, Latvian, Lithuanian, Norwegian, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Spanish, Spanish (Latin America), Swedish, Turkish and Ukranian.
How long does it take to process a media, what's the turn-around-time?
Processing a video in Fluen Studio consists of three key steps: media encoding, caption generation, and optional translation generation.
Media encoding, influenced by the file's length and size, typically lasts between 30 seconds and a few minutes. Caption generation, also dependent on the media's duration, may take up to a few minutes. Translation generation commences once the caption generation is complete and the caption review/post-edit has been completed, if you opted-in for the additional post-editing step. If multiple languages are required, translation processes occur simultaneously, usually completed in minutes.
Overall, you can estimate the total TAT (turn-around-time) to be less than half the duration of the video itself. The estimates vary with file size and system load.
Exporting the video with hardcoded subtitles is a separate process and may take a bit longer. Typically, consider anything between 1/4th and half of the video duration. The time it takes depends on the bitrate used in the original file (the higher the bitrate, the longer it'll take to process) and wether the export is for a 4k video, 1080 or 720p - lower resolutions should take less time.
Media encoding, influenced by the file's length and size, typically lasts between 30 seconds and a few minutes. Caption generation, also dependent on the media's duration, may take up to a few minutes. Translation generation commences once the caption generation is complete and the caption review/post-edit has been completed, if you opted-in for the additional post-editing step. If multiple languages are required, translation processes occur simultaneously, usually completed in minutes.
Overall, you can estimate the total TAT (turn-around-time) to be less than half the duration of the video itself. The estimates vary with file size and system load.
Exporting the video with hardcoded subtitles is a separate process and may take a bit longer. Typically, consider anything between 1/4th and half of the video duration. The time it takes depends on the bitrate used in the original file (the higher the bitrate, the longer it'll take to process) and wether the export is for a 4k video, 1080 or 720p - lower resolutions should take less time.
How do I post-edit a track?
Post-editing a track in Fluen Studio is straightforward. From your dashboard, look for the media track highlighted in blue, indicating it's in the 'Pending Review' status. Click on it to access the subtitle editing tool. This tool enables you to view all subtitle cues, edit text, cut/copy/paste content between cues, and adjust timestamps by moving the handles in the waveform. The best part? Any changes you make are autosaved instantly, and you can review them in real-time on the media player.
Once you're done with your changes, click on the "Complete" button on the top-right of the subtitle editor screen, to move the process into the next step in the workflow.
You can choose whether to post-edit a track or not during the media upload process, or set it as your default behavior in your account settings. You can also edit your tracks once they're already completed. Whichever method suits your workflow best.
Once you're done with your changes, click on the "Complete" button on the top-right of the subtitle editor screen, to move the process into the next step in the workflow.
You can choose whether to post-edit a track or not during the media upload process, or set it as your default behavior in your account settings. You can also edit your tracks once they're already completed. Whichever method suits your workflow best.
Do I have to manually post-edit every single subtitle track?
Not at all. In fact, the auto-generated subtitles provided by Fluen Studio are typically of high quality and will meet your needs in most instances. However, if you plan on translating the captions, we do recommend reviewing and post-editing the source language captions. Doing so can enhance the accuracy and overall quality of the auto-generated translation tracks, ensuring a better final product.
How can I fine-tune the timestamps of a subtitle cue?
Fine-tuning subtitle cue timestamps is simple with Fluen Studio. On the subtitle editor screen, each subtitle is highlighted on the waveform at the bottom. To adjust a subtitle's timing, click and drag the edges of the highlighted area on the waveform to your preferred position. However, it's important to note that our automatic spotting and subtitle alignment should already provide highly accurate timings, so extensive adjustments may not be necessary.
Why isn't there a one-to-one correspondence between caption and translation cues?
Fluen Studio prioritizes the quality and readability of translations, leading us to adopt a unique approach. Unlike maintaining a strict one-to-one correspondence between caption and translation cues, we choose to group source captions into smaller paragraphs for translation. This approach offers a more coherent and contextually accurate translation output.
After translation, this text is then re-segmented into the target cues. The re-segmentation maintains correct timestamps from the source or generates new ones at appropriate break points, even if those break points did not exist in the source cues. This way, we ensure that the translated subtitles are not just accurate, but also fluid and easily understandable, enhancing the viewer's experience.
After translation, this text is then re-segmented into the target cues. The re-segmentation maintains correct timestamps from the source or generates new ones at appropriate break points, even if those break points did not exist in the source cues. This way, we ensure that the translated subtitles are not just accurate, but also fluid and easily understandable, enhancing the viewer's experience.
How does Auto-Segmentation work?
Auto-Segmentation in Fluen Studio uses advanced Natural Language Processing (NLP) to interpret the grammatical structure of each sentence in a language-specific way. This means the process takes into account the unique grammatical and stylistic conventions of each supported language to determine optimal endpoints for each subtitle cue and the best places to insert line breaks.
This approach mirrors the strategies used by professional linguists when creating subtitles from scratch. They consider a sentence's grammatical structure and apply standard line treatment rules such as breaking lines after punctuation marks or before conjunctions and prepositions, while avoiding separating closely related elements like a noun from its accompanying article or adjective, first names from last names, or verbs from their subject pronouns.
Our auto-segmentation algorithm integrates these well-established subtitling rules, all the while considering other constraints such as the maximum characters per line. The result is a thoughtful blend of linguistic insight and technical precision to deliver high-quality subtitle segmentation.
This approach mirrors the strategies used by professional linguists when creating subtitles from scratch. They consider a sentence's grammatical structure and apply standard line treatment rules such as breaking lines after punctuation marks or before conjunctions and prepositions, while avoiding separating closely related elements like a noun from its accompanying article or adjective, first names from last names, or verbs from their subject pronouns.
Our auto-segmentation algorithm integrates these well-established subtitling rules, all the while considering other constraints such as the maximum characters per line. The result is a thoughtful blend of linguistic insight and technical precision to deliver high-quality subtitle segmentation.
What do you mean by Auto-Wordsmith? Is it just for translations?
Auto-Wordsmith is an innovative capability enabled by AI and large language models (LLMs). It emulates a crucial aspect of professional linguists' work during subtitle translation: the skillful rewriting and adaptation of translated cues to meet character-per-second constraints or ensure optimal readability.
In certain language pairs, such as English to a more verbose language like German, the translated text may exceed the available time for display or become overly convoluted. In all cases where the translated text greatly exceeds the length of its source, our AI-driven Auto-wordsmith feature automatically reformats and rewrites the translated segments while preserving the original meaning. By eliminating unnecessary content or refining the wording, the translated segments become more digestible and readable for viewers without sacrificing the core message.
Auto-wordsmith is only applicable to auto-generated translations rather than captions. Since captions aim to provide a verbatim representation of the source content, they typically contain more literal and unchanged text.
In certain language pairs, such as English to a more verbose language like German, the translated text may exceed the available time for display or become overly convoluted. In all cases where the translated text greatly exceeds the length of its source, our AI-driven Auto-wordsmith feature automatically reformats and rewrites the translated segments while preserving the original meaning. By eliminating unnecessary content or refining the wording, the translated segments become more digestible and readable for viewers without sacrificing the core message.
Auto-wordsmith is only applicable to auto-generated translations rather than captions. Since captions aim to provide a verbatim representation of the source content, they typically contain more literal and unchanged text.