MMAudio is a CVPR 2025 project for video-to-audio synthesis that generates synchronized audio from video and/or text inputs. It uses multimodal joint training and a synchronization module to align generated audio with video frames. A pretrained large_44k_v2 model is used by default, with inference around 6GB of GPU memory in 16-bit mode.
All Voice Lab offers AI voice cloning, text-to-speech and a real-time voice changer for creators, aimed at video dubbing and podcasts. Its text-to-speech is described as adapting tone, rhythm and pitch to text sentiment, and the site states the AI integrates 33 major languages. A proprietary voice model named MaskGCT is also presented, alongside API and MCP access.
Fish Audio provides AI text-to-speech and voice cloning with emotion control and multilingual support. It offers TTS, speech-to-text, voice agents, and developer APIs with Python and TypeScript SDKs. The site lists a free tier with 8,000 monthly credits and paid Plus, Pro, and Max plans.
MMAudio is a research project on multimodal joint training for video-to-audio synthesis. It generates synchronized audio given video and/or text inputs. The work is presented at CVPR 2025 by authors from the University of Illinois Urbana-Champaign and Sony.
Google Flow Music is a generative AI platform for creating, remixing, and sharing songs. It includes a chat with Producer, AI music video, and tools to build audio plugins, music games, and custom DAWs. It is free to start without a credit card.
FreeTTS is a browser-based audio workspace for creating speech, transcribing recordings, isolating vocals, enhancing voices, and editing audio. It offers Device TTS, AI Local TTS, and AI Cloud voices with free monthly characters. Browser tools handle audio cutting, joining, compression, and conversion, and a free plan includes 5,000 characters per month.
MultiSong Leaderboard is a ranking page on the MVSEP music and voice separation platform. It evaluates multisong audio separation algorithms using metrics including SDR, SI-SDR, L1 frequency, log WMSE, aura STFT, aura MRSTFT, bleedless, and fullness, with rankings updated every three hours. The page lists top-performing algorithms and their scores for separation tasks such as bass, drums, other stems, vocals, and instrumentals.
Community notes
Be the first to share how this project worked for you.
Sign in with a verified account to leave a review or join the discussion.
Sign in to reviewKeep exploring
All Voice Lab offers AI voice cloning, text-to-speech and a real-time voice changer for creators, aimed at video dubbing and podcasts. Its text-to-speech is described as adapting tone, rhythm and pitch to text sentiment, and the site states the AI integrates 33 major languages. A proprietary voice model named MaskGCT is also presented, alongside API and MCP access.
Fish Audio provides AI text-to-speech and voice cloning with emotion control and multilingual support. It offers TTS, speech-to-text, voice agents, and developer APIs with Python and TypeScript SDKs. The site lists a free tier with 8,000 monthly credits and paid Plus, Pro, and Max plans.
MMAudio is a research project on multimodal joint training for video-to-audio synthesis. It generates synchronized audio given video and/or text inputs. The work is presented at CVPR 2025 by authors from the University of Illinois Urbana-Champaign and Sony.
Google Flow Music is a generative AI platform for creating, remixing, and sharing songs. It includes a chat with Producer, AI music video, and tools to build audio plugins, music games, and custom DAWs. It is free to start without a credit card.
FreeTTS is a browser-based audio workspace for creating speech, transcribing recordings, isolating vocals, enhancing voices, and editing audio. It offers Device TTS, AI Local TTS, and AI Cloud voices with free monthly characters. Browser tools handle audio cutting, joining, compression, and conversion, and a free plan includes 5,000 characters per month.
MultiSong Leaderboard is a ranking page on the MVSEP music and voice separation platform. It evaluates multisong audio separation algorithms using metrics including SDR, SI-SDR, L1 frequency, log WMSE, aura STFT, aura MRSTFT, bleedless, and fullness, with rankings updated every three hours. The page lists top-performing algorithms and their scores for separation tasks such as bass, drums, other stems, vocals, and instrumentals.
