We have found that radio and television commercials, corporate and sales videos, episodes on YouTube and television, or movies each require their own approaches. Audio production cannot be done the same way for each format. But is it only the case with audio editing? Absolutely not. The challenges to the entire team are different for each format.
The locations and sets, the shooting style, the actors are also different each time. What’s fast? Short commercials can be shot, edited, mixed, and released relatively quickly. Radio spots? Even faster since there’s no picture involved. But long feature films and television formats require much more effort. Filming locations vary widely. Lighting is the most complex for feature films. And the sound? Since you can’t see the sound, most people don’t realize that microphones pick up everything, really everything, until afterwards when they evaluate the footage. The brain blocks out certain sounds that don’t belong to what you’re imagining.
So, noise or technical defects that the recording technician didn’t hear sometimes have to be corrected or reworked in a painstaking and time-consuming way. In the worst case, it is necessary to make a new recording.
That’s when ADR comes into play. The easiest task is when the person seen in the picture only re-records individual parts of the text that cannot be used due to a bad sound recording.
It becomes more complex to dub the entire film into another target language. An important and, above all, cost- and time-saving prerequisite for this is that the dialogues can be replaced independently of the rest of the sounds, such as foley, atmospheres and music.
From this moment on, we, blk-germandubbing, come into question as a possible partner for you.
First, we start with the rough translation of the dialogues from the original film. Already at this point the course is set for more or less costs and time. Why?
Is there even a written original of the texts? Does the text in the document really match the spoken text in the film? Or does the entire film has to be transcribed first before the raw translation? Well, transcription is already done relatively well automatically today, thanks to AI. Nevertheless, this result must first be turned into a readable original through post-editing.
Secondly, phrases that cannot be translated one-to-one are explained in notes. Especially colloquial expressions, idioms, metaphors, and idioms must be explained in the so-called “subnotes”. This contributes to a better understanding of the statement and shows possibilities how it could be dealt with in the target language.
Especially subject-specific terminology should be researched thoroughly and rendered as adequately as possible. This applies, for example, to films or series from medical, technical, and legal fields.
Especially when dialogues refer to typical national social events, institutions or prominent personalities that are not known abroad and these allusions are not comprehensible to the target audience, detailed background information is required. If, for example, a reference is made to an American television classic that is completely unknown in the target language area, an equivalent should be found that corresponds to it.
Before working on the dialogue, dubbing writers must recognize the linguistic level of the protagonists and analyze their cultural background and attitude to life, so that within this framework the individual dialogues can be formulated more freely if necessary. The task is essentially to find a new, meaningful text that is true to the scene, in sync with the mouth movements captured in the image and the given text content. All passages that appear asynchronous must be avoided, slang expressions and puns of the target language must be found, or linguistic unevenness must be removed by editing the text.
Likewise, one must subordinate oneself above all to the mouth movements of the actors (in the “on”) that can be seen in close-ups. There, sentence lengths and the position of certain consonants that cause lip closures, so-called labials, are predetermined and the translation must be adapted true to meaning according to these specifications. The text does not have to correspond to the original text, only the original mouth movements are important.
Consonants, such as “b”, “p” and “m” are clearly recognizable since their production can be read from the lips. The sounds “v”, “w” and “f” in German are formed very similarly, so that they can be interchanged. Fellow sounds such as “d”, “t”, “k”, “g”, “s”, “r” are also interchangeable because they are pronounced without specific lip position. The articulation of vowels, however, are important because they are represented by extreme lip or jaw openings.
The dubbing script must be coherent in every respect and written in a good and fluent language. It fits the language level of the acting characters, the facial expressions, and gestures, can be lip-synced and reproduces the content of the original as precisely as possible.
The final timecode-based dialog book is formatted in both languages in table format as a CSV file in preparation for import into audio software, such as Nuendo. In the process, each marker can additionally contain various user-defined details, such as timecode postion (cue-point), the name of the role or the location of the scene. Markers can be exported, giving the editor maximum freedom in organizing the project.
Based on these specifications, the dubbing writer adapts the original text line by line. This is done either by following the subdivisions created by the editor into short individual sections, so-called takes, which usually comprise only one or two sentences and are about two to ten seconds long. In the final version of the dubbing book, all the dialogue to be spoken is given in takes with consecutive numbers.
Shorter or longer pauses in the dialogue and whether the character is seen in the picture from the front, only from behind, or not at all while speaking are also noted. These notes, in parentheses, contain indications such as (sound), (Procrastination), (into ON), (into CONT), (into OFF) or suggest an alternative text (2. version:)
The number of takes is counted as accurately as possible for a complete list. This includes what roles there are and how many takes each role has. Here, a relatively intuitive decision must often be made as to what constitutes a take. Longer sentences or passages are often subdivided, sometimes very finely, according to dramaturgical or scenic aspects.
The previously created take list is passed on to the dispatchers or recording supervisors. Based on the film, a brief description of the role and the number of takes, they decide which dubbing actor should speak which role. Often, voices are cast that also have a visual similarity to the person seen in the picture. Several voice actors are always considered here so that one has fallback options in case someone falls ill or cannot participate at the specified times for other reasons.
Casting is an essential aspect of an optimized production concept. An old directorial saying goes, “Casting the right talent is half the “rent”!”
The disposition is not based on chronology, but on persons. The scheduling therefore takes into account the individual parameters of the actors. Leading roles are often booked by the day, while supporting roles have to sit in the waiting area until it’s their turn for one to four takes. Equally important is the estimation of the time required. Is the voice actor known for always needing 4-5 takes until the final recording or is the text filled with difficult technical terms or do many roles have few takes, which extends the time scheduling for the change of characters.
At the agreed time, the voice actor appears in the recording studio, also called a dubbing studio. Usually there are two separate rooms, which allow the director, the editor, and the sound engineer to control the distance between the speaker and the microphone through a large glass pane. The actual synchronicity is controlled on a separate screen. In the recording room there will be the microphone and a large screen on which all the necessary information (an in-counter, the film, the text, status of the recording) will be displayed. The insert lists are already installed in the audio software along with the video in the computer, and by clicking on the first take for the cast role, everything is now in place (cue point). To get started, the original sound is often heard first to get an idea of the character and pitch in the original.
If the take is vocally correct and in sync, the next take (cue point) is activated, and everything starts again from the beginning. A good voice talent can do a take in 2-3 passes on average. The full professionals try to do it in one pass. In Germany, this is called an “ace”, like in tennis.
Once all the recordings have been made, each individual position is checked again by the editor for synchronicity and an individual audio track is created for each cast. These are aligned with each other in a final mastering process and made available in the desired format to the customer or, if the mixing is to be done at blk-germandubbing, to the mixing engineer via data transfer.
We hope that these brief descriptions have given you an overview of the main processes involved in localizing your content. My experience as a sound engineer, producer and voice-over artist goes back to 1986 and I am glad that nowadays we don’t have to deal with analog tapes and U-Matic video cassettes anymore.
We look forward to working with you to bring your original videos and films into the German language.