Rapid Voice Cloning 2.0: New Voice Cloning Model with Unmatched Accuracy
Blog post from Resemble AI
Resemble AI has launched Rapid Voice Clone 2.0, a breakthrough text-to-speech model that allows users to generate high-quality voice content using just 20 seconds of audio, with capabilities for seamless voice editing and localization. This technology offers precise control over voice attributes, ensuring authentic and engaging experiences across diverse applications like gaming, media, and personalized messaging. A comparison with competitor ElevenLabs demonstrated Resemble AI's superior ability to retain prosodic attributes and natural accents, with 85% of survey respondents favoring its output. The model's effectiveness is highlighted through its integration in projects like Red Games' Crayola Adventures and TrueFan, which leveraged the technology for personalized interactive experiences and celebrity messages, respectively. Now available on Google Cloud Marketplace, Resemble AI can scale its services globally, offering enhanced speed and accuracy through Google Cloud's infrastructure. The company remains committed to ethical AI use, emphasizing security through features like deepfake detection and neural watermarking, ensuring content integrity while advancing the capabilities of generative AI.