Can AI dubbing achieve human parity? With the launch of DupDub(www.dupdub.com), the answer is definitely yes. Recently, Mobvoi has launched the fifth-generation TTS engine MeetHiFiVoice, which has qualitatively improved the sound quality and rhythm, and pushed the AI ​​dubbing effect "comparable to real people" to a new extreme height again.

After the release of our first-generation TTS engine in 2015, Mobvoi achieved a new technological breakthrough again, and launched the industry-leading fourth-generation speech synthesis system MeetVoice (short for Mobvoi End-to-End TTS Voice) in August, 2019. Since its release, MeetVoice has been deeply applied to Mobvoi's various product lines and various ToB manufacturers. The well-known AI dubbing artifact - "Moyin" (Chinese version of DupDub, official website at www.moyin.com ) has launched hundreds of voices on a large scale, which is well received by users.

With consistent iterations, the effect of our speech synthesis system has been constantly approaching "human parity". In the past three years, with the rapid development of the short video industry, our users have put forward higher and higher demands on the synthesis effect. Mobvoi team has continued to upgrade the MeetVoice model, including Prosody Adjustment, 48K Studio Quality, Stress and Voice Conversion are widely loved by users.