Appendix Section
Examples of Generative AI models
These illustrations show the range of uses and domains for which generative AI is appropriate. Not only do the models demonstrate impressive capabilities, but they also highlight the necessity of a more thorough investigation of the moral dilemmas, possible biases, and the critical requirement for responsible implementation that exist in both the developmental and operational stages of generative AI systems.As we work throughout the broad field of generative AI, we need to be mindful of the subtle nuances that come with using such strong technology, and we must carefully address ethical concerns. It entails a thorough examination of any biases that might unintentionally arise during deployment or training, and it encourages researchers and practitioners to implement safety precautions to prevent undesirable results.Furthermore, ethical use of generative AI is necessary, which means that any possible harm to society must be kept to a minimum. This suggests that these systems need to uphold moral principles and advance the goals of the greater technological community. This is a turning point in the responsible development of AI, since it highlights how important it is to integrate generative AI with care and purpose given the intricate relationship between technical progress and societal wellness.
| GENRATIVE MODELS | SOFTWARE TOOL NAME |
|---|---|
| Text to Image (T2I) | DALLE-E 2 Stable Diffusion Craiyon Jasper Imagen MidJourney NightCafe GauGAN 2 Wombo Wonder neural.love Pixray-test2image |
| Text to Video (T2V) | Runway Gen-2 ModelScope ZeroScope VideoCrafter Synthesis Kaiber Wonder Studio Phenaki Meta’s Make-A-Video Nvidia’s Latent Diffusion Model |
| Text to Audio (T2A) | Murf.ai Play.ht Resemble.ai WellSaid Descript lovo.ai Speechify Listnr Sonantic Woord |
| Text to Text (T2T) | Simplified Frase Requstory Grammarly Market Muse HubSpot Flowrite SudoWrite Copysmith Ideasbyai |
| Text to Motion (T2M) | MDM: Human Motion Diffusion Model TREEInd. VQGAN-CLIP |
| Text to Code (T2C) | StarCoder OpenAI Codex GitHub Copilot CodeT5 Polycoder Replit Ghostwriter Tabine |
| Text to NFT (T2N) | ArtBreeder DeepDreamGenerator Deep Art Effect StyleGAN RunwayML Google Muse AI Prisma NeuralStyle AI |
| Text to 3D (T2D) | DreamFusion Clip - Mesh GET 3D Mochi Masterpiece Studio Spline AI Meshcapade |
Table 3: Examples of Generative AI models by taking Text data as Source
| GENRATIVE MODELS | SOFTWARE TOOL NAME |
|---|---|
| Audio to Text (A2T) | 1.WaveNet 2.DeepVoice 3.tacotron 4.MelGAN 5.hiFiGAN 6.Descript 7.AssemblyAI 8.Whisper(OpenAI) |
| Audio to Video (A2V) | 1.Audio2Vec 2.MusicVAE 3.MoCoGAN 4.VCGAN 5.Diffusion models |
| Audio to Audio (A2A) | 1.AudioLM 2.VOICEMOD |
Table 4: Examples of Generative AI models by taking Audio as Source