A recent court ruling in Shanghai has brought attention to the legal boundaries surrounding AI-synthesized voices and the protection of natural persons' voice rights.

According to reports, the Shanghai First Intermediate People's Court has concluded a case involving a dispute over the protection of a natural person's voice rights triggered by AI-synthesized speech, ruling that the platform operator committed voice infringement and ordering it to pay 50,000 yuan in compensation for the rights holder's economic losses.

Background of the Case

Ms. Wang is a voice actress who completed real-name authentication on a certain app, where she regularly performs dubbing work and publishes her recordings to earn income.

In January 2025, Ms. Wang learned from a friend that Company A was suspected of using her voice in an internet promotional campaign aimed at acquiring new users.

After verification, Ms. Wang believed that the voice used in the campaign was likely AI-synthesized from her own voice, so she notarized the relevant audio to preserve evidence and filed a lawsuit with the people's court.

Ms. Wang argued that she had never cooperated with or authorized Company A in any way, and that the company's unauthorized use of her voice constituted infringement. She demanded that Company A cease the infringement and pay 300,000 yuan in compensation for economic losses.

Company A's Defense

Company A argued that the audio in question was indeed generated by AI developed by the company, but the specific source and the materials fed into the system could not be confirmed because the former employee responsible had left the company.

The company claimed it had previously been unaware of Ms. Wang's identity as a voice actress and that there was no collection or use of her voice. It also argued that the existing evidence could not directly prove that Ms. Wang's voice was used in its promotional activities.

Judicial Appraisal and Key Evidence

During the trial, to prove the facts of the infringement, Ms. Wang conducted a voice analysis comparison between the notarized audio in question and her own voice recorded on the spot through judicial appraisal.

The appraisal opinion showed that among 28 formant acoustic indicators, 24 indicators in the audio in question had a deviation of less than 10% from Ms. Wang's own audio, with 16 of them having a deviation of less than 5.36%. The portions that were relatively similar and highly similar reached 90%.

The court found that Company A's actions constituted AI-generated voice infringement and caused economic losses to Ms. Wang, and accordingly ruled that Company A should compensate Ms. Wang 50,000 yuan.

Both parties were dissatisfied with the first-instance judgment and filed appeals.

Appellate Court's Reasoning

After trial, the Shanghai First Intermediate People's Court held that using a natural person's voice as training corpus without their consent to imitate that person's timbre, intonation, and pronunciation style to generate a synthesized voice that can identify that natural person should be recognized as infringing upon the natural person's voice rights.

In this case, Ms. Wang, as a voice actress, had already provided evidence proving that her dubbing works were published on public online platforms, and Company A had information channels and possibilities to obtain the source of her voice.

Meanwhile, according to the appraisal report analysis, the synthesized voice and the human voice audio had a small degree of deviation and were highly similar, sufficient to enable ordinary members of the public to associate them with a specific identity, possessing identifiability.

Although Company A denied the allegations, as the party in control of the original training data, it failed to fully provide evidence on facts to be proven, including the legality of the source of the fed materials, the non-use of natural person voice characteristics, and the absence of public confusion or misidentification. It should therefore bear the adverse consequences of failing to meet the burden of proof.

Accordingly, it could be determined that Company A, without Ms. Wang's consent, commercially used a synthesized voice capable of identifying an individual's identity to the public, which constituted voice infringement.

Determination of Damages

Regarding the determination of the amount of damages for voice infringement, the court held that priority should be given to determining the amount by reference to the rights holder's licensing fees for similar uses. If the rights holder's evidence is insufficient and the licensing fees are difficult to verify, the amount may be comprehensively determined by considering factors such as the duration of the infringement, the scope of dissemination, the manner of use, the degree of fault, the results of damage, and the proportion of contribution of personality elements in commercial exploitation.

In this case, the voice in question was available for a relatively short period of time and was used as a process broadcast prompt sound effect, not as a core selling point or traffic-driving vehicle for the campaign. In addition, Ms. Wang's online visibility and public activity were not high, and the independent commercial market value of her voice was limited. Therefore, the court found that the first-instance court's discretionary award of 50,000 yuan in compensation was fair and reasonable.

Final Judgment

In summary, the Shanghai First Intermediate People's Court ruled in accordance with the law: the appeal is dismissed and the original judgment is upheld.