German voice actor, Manfred Lehmann, best known as the German dubbing artist for
Bruce Willis, among other audiobook and radio play narration, recently successfully
sued a YouTube channel operator for unauthorized commercial exploitation of his voice in the Berlin Regional Court. Described as
'deep, multi-faceted, powerful, very manly, very striking', Lehmann’s voice was imitated by AI and used to provide commentary in two videos uploaded by a YouTube creator with 190,000 subscribers and an online shop. The YouTuber’s defence? That they ‘bought’ the AI voice and thus had ‘rights of use’, particularly for satire purposes.
 |
Photo by Vidak on Unsplash Perhaps its time for this Kat to 'Claw Hard'
|
For Lehmann, this had serious public reputational damage. The YouTube channel was ‘right-wing’ leaning and its online shop sold t-shirts with the slogan “Woke Zero”. It also comes a few months after
German voice actors, studios, and unions galvanised against the use of AI on the ‘dubbing and localization sector’ given its impact on artistic livelihoods and artistic integrity. In Germany, we can think of
the 'Oscars of Dubbing', and across the EU,
voice actors find it akin to stealing their identity. And so while the Netherlands and Denmark might
expand the contours of protected subject matter, Lehmann’s case, perhaps, more closely aligns with the actual harm, by asserting that the use interferes with the German general right of personality found in
Article 2 of the German Basic Law (Germany’s constitution), even though the Court did not expressly refer to it.
The general personality right extends to your voice
The general right of personality, specific to voice, is not well-defined in German jurisprudence. In comparison to the right to protect one’s image, codified in section 22 of the
German Act on Copyright in Artistic Works and Photography, protection over one’s voice is not regulated by statute. Its scope is a little ambiguous, with the court reflecting that some scholars view it as an analogous structure to section 22, while others assume protection more broadly through the general personality right. Regardless of the exact status of the copyright nexus, the personality right’s scope was affirmed by the German Federal Court of Justice (BGH) in
Marlene Dietrich to include the commercial use of a person’s distinctive personality characteristics that have considerable economic value.
In the present case, the Court found that the YouTube creator had infringed the general personality right by using an AI-generated voice to add narration to videos. While it was not Lehmann’s voice, but an artificial version, the method of production was irrelevant. What matters is whether a ‘not insignificant’ portion of the target audience, given the similarity between Lehmann’s voice and the AI-generated voice, will assume that Lehmann has authorised the commentary. The Court explains this test in (almost) ‘trade mark’-like terms. The decisive factor is the confusion caused by the deliberately created similarity of the voices which could lead viewers to believe that Lehmann consented to the use. The Court pointed towards YouTube users commenting on the video pages that it was Lehmann’s voice as evidence of such confusion.
Lawful creative reuse or unauthorised commercial reuse?
Interference with the general personality right invites fundamental rights balancing. And so the question became whether the YouTube creator’s freedom of art and opinion justified the reuse. The YouTube creator argued that these were satirical videos and that he chose an authentic voice with ‘a heroic sound’. Further, as it was a synthetic imitation suggested by AI software, it was not the original spoken voice, but an AI voice similar to Lehmann’s pitch.
Despite these factors, the Court found that the reuse primarily served the defendant’s commercial interests. The videos may have had satirical content, as the creator is mocking the incompetence of the government at the time, but they lacked focus on Lehmann’s or other people’s behaviour (or voice). Rather the YouTube creator’s intent was to make the videos more appealing and attract as many internet users as possible. The Court pointed towards the online shop, linked at the end of each video, to demonstrate this commercial benefit. They found that the use’s primary objective was commercial as the use of Lehmann’s voice was to increase click-through rates and sales.
Freedom of art and opinion did not outweigh Lehmann’s interests over his voice dues to its commercial nature. The YouTube creator’s expression was not restricted by Lehmann's rights as he could have paid for a licence. If voice rights are structurally similar to section 22 of the German Act on Copyright in Artistic Works and Photography, the court found that the reuse remains unlawful. There is no reference to a contemporary historical event, satire, or art. The fact that a sizeable portion of the public would conclude that the use ‘represents actual statements by the cloned individual’ extinguishes a claim based on artistic freedom. The section 22 analogy would also strengthen Lehmann’s interest to be paid for the commercial use of his voice as an
'adaptation or transformation'.
The right-wing leaning reputation of the YouTube channel, in combination with the products sold at the online shop, presented a huge risk for Lehmann’s public reputation, and was considered a serious factor. Similarly to
Helene Fischer, it appears that reputational-based harm is easier to prove for right-wing leaning reuse. Indeed,
some have questioned whether there can be ‘right-wing satire at all’.
Drawing upon BGH jurisprudence on the unauthorized use of images for advertising purposes, the Court explained that compensation must commensurate with the value of the use. Given Lehmann’s ‘sought-after’ status in the German advertising industry, the starting price would be €1,800 for limited commercial use of his voice. However, the size of the YouTube channel, the unlimited nature of the use, alongside Lehmann’s very distinctive and well-known voice, resulted in a license fee of € 2,000 per video clip.
Comment
Lehmann presents a significant first step within the generative AI creativity landscape. While a first instance decision, the extension of the general personality right over voice provides an alternative framework for filling gaps that have seriously challenged copyright law. Though the Court evades discussion on the nexus between the general personality right and the copyright system, strengthening the right responds to the inherent human value connected to data used to train AI models - one that most certainly has commercial value.
This is a more wholistic and flexible approach that balances numerous rights, including personal data under article 6
GDPR, and reaches a proportionate outcome. It achieves something that the AI Act fails despite its broad definition of a
deep fake and provisions on
labelling AI-generated content. Indeed the Court mentions in obiter that the YouTube creator gave no indication on either video that the voice was AI-generated.
Overall this is an incredibly timely development as we await the CJEU
Pelham II reference on pastiche (IPKat
here). Advocate General Emilio was quite clear on the limits of pastiche. Perhaps it is time that we begin to look outside the copyright law box.