A How-To Guide to Suno

Whether crafting a song for a loved one, enhancing your video projects with a custom soundtrack, or simply experimenting with your musical ideas, Suno offers an intuitive way to bring your creativity to life. Here’s everything you need to know with this powerful AI tool.

What Can You Do with Suno?

Suno makes music creation accessible for everyone, regardless of skill level. Here’s how you can use it:

• Generate music with a single prompt: Describe your idea, and Suno will create a unique tune based on your input.

• Get inspired with the AI Lyric Generator: Quickly generate lyrics to kickstart your songwriting process.

• Customize your music: You can tailor your creations by tweaking lyrics, adding metatags, choosing a music style, and even mimicking the tone of other tracks created within Suno.

Understanding Suno’s Effectiveness

AI music generation is an exciting frontier, but like all creative tools, it has nuances.

Suno excels in making music creation fast, accessible, and enjoyable. If you're new to music production or need a quick solution, it’s an excellent tool for:

• Idea generation: Quickly explore different musical concepts without needing advanced technical skills.

• Non-professional projects: Create engaging tracks for personal or casual use, like videos, gifts, or experiments.

Suno requires iteration and experimentation for the best results, as with any other AI tool. As with AI-generated text, images, or videos, you’ll rarely accept the first output for complex or highly specialized tasks.

Also, similarly to using AI for poetry or any other creative field, to receive a professional-quality result, AI output needs a human touch. The more sophisticated or specific your project, the more you’ll need to guide and mold the output.

Who Is Suno Best For?

• Casual creators: Suno is perfect for generating quick, fun results if you’re not a professional musician but want a simple music-making method.

• Professionals and hobbyists: Suno can be a great starting point for brainstorming ideas, which you can refine using traditional production methods.

Suno’s Features

To start with Suno, go to their website and create a free account. Then, choose from one of these options.

1. AI music with one prompt

• Write your first prompt based on a theme, mood, or story you’d like your song to convey. Suno allows you to explore new styles of music for inspiration.

• Click “Create.”

• Suno will output two versions of the music/song created based on your prompt.

Tips for Suno prompts:

• Separate genre, voice, instrumental emphasis, and other attributes with commas

• Use brackets to define song sections or vocal styles

• Switch to the custom mode described below for more precise results

2. AI music with advanced customization

• Craft your lyrics: You can use Suno’s AI Lyric Generator or an external LLM like Claude or ChatGPT.

• Metatags: You can optionally add metatags such as [Chorus], [Verse], or [Break] to tell Suno how to handle specific parts of your song.

• Style and genre: Describe your desired musical genre and style here.

• Persona (Beta): Personas allow you to capture the vocals, energy, and atmosphere of an existing track and use it to create a brand-new one. To create a persona:

  • Find a song you love

  • Click the “More Options” icon above where the lyrics are displayed

  • Select Create -> Make a persona

• Title: Give your song a title to make it easy to store, find, and organize.

• Upload your music: You can also upload your audio to Suno between 6 and 60 seconds long.

Customization options:

• Instruments: Include specific instrumental preferences

• Mood: Adjust the vibe with descriptors like “upbeat”, “melancholic,” and “energetic”

• Vocal style: Specify vocal techniques for a unique sound

• Technical musical elements: Suggest chord progressions, musical keys, tempo or rhythmic patterns

Hope it was helpful! Read about 5 unique use cases in our latest blog post.

Latest AI News

Practical AI Tools

• Microsoft’s Copilot Relaunch: Microsoft has reintroduced Copilot, a productivity-focused AI assistant, for $3/month on top of its existing 365 plans. Users can choose between AI integration and a “classical” non-AI setup.

• Google’s Gemini in Workspace: Google bundled its Gemini AI into all Workspace plans, increasing costs by $2/month. The integration covers tools like Google Docs, Sheets, and other apps inside Google Workspace. Users can’t opt out of these AI enhancements showing Google’s broader push to make AI a standard part of office tools.

• Luma AI’s Ray2: Ray2 is Luma AI’s latest tool for creating short, visually stunning video clips (5–10 seconds). The model uses advanced AI to generate smooth, cinematic scenes perfect for ads and media. Key features include the ability to understand natural language instructions and ensure character consistency.

• Runway’s Frames: Runway has introduced Frames, a cutting-edge tool for generating high-quality images with precise style control. Frames excels at stylistic consistency and integrates with Runway’s video tools, meaning users can transition from static visuals to dynamic video projects effortlessly.

• OpenAI’s Operator: OpenAI has finally rolled out Operator, an advanced AI assistant designed to handle real-world tasks like booking vacations, filling out forms, or ordering groceries. Currently, it’s available to ChatGPT Pro users, with plans for broader access soon.

• Anthropic’s $1 Billion Funding: Anthropic has secured $1 billion in funding aimed at upgrading its Claude AI model. Future improvements include two-way voice interaction and enhanced memory, allowing Claude to hold longer and more detailed conversations.

For Developers

• DeepSeek-R1: DeepSeek, a Chinese AI startup, has released DeepSeek-R1, an open-source reasoning model performing on par with, and occasionally outperforming, OpenAI's o1 advanced reasoning model. This open-access approach challenges the closed nature of many U.S.-developed models.

• Google Titans: Google has introduced Titans, an AI architecture that brings long-term memory capabilities to language models. Titans can process and retain information across contexts exceeding 2 million tokens. The model dynamically updates its memory as it processes new inputs, allowing it to "learn on the fly" without retraining.

• RAG 2.0 from Retrieval-Augmented Generation Creators: Contextual AI has launched a platform for creating end-to-end retrieval-augmented generation (RAG) systems. RAG models combine the power of AI-generated responses with external knowledge retrieval to ensure accuracy and groundedness.

• X-CLR: Researchers have introduced X-Sample Contrastive Loss (X-CLR), an upgraded approach to contrastive learning. It uses continuous similarity scores instead of discrete values like "similar" or "not similar." It’s an important development in self-supervised learning, enabling models to better understand real-world complexities.

• NVIDIA's Personal AI Supercomputer: NVIDIA is making powerful AI development tools more accessible with Project DIGITS, a desktop AI supercomputer set to launch in May 2025. Starting at $3,000, it’s powered by the new GB Grace Black Superchip, which supports models with up to 200 billion parameters, a capability usually reserved for massive data centers.

AI and Science

• Brain-Computer Interface (BCI): A 69-year-old paralyzed man controlled a virtual drone using only his thoughts. This was made possible through a brain-computer interface (BCI) developed by researchers from the University of Michigan and Stanford University. By imagining finger movements, the participant could send commands to control the drone.

• DeepMind’s AI-Designed Drugs: DeepMind and its spinoff company, Isomorphic Labs, are making waves in pharmaceutical research. According to CEO Demis Hassabis, the first AI-designed drugs are expected to enter clinical trials by the end of 2025. AlphaFold predicted the structures of over 200 million proteins, helping accelerate drug discovery and reducing the timeline from 10+ years to just weeks or months.

• OpenAI’s GPT-4b: OpenAI has collaborated with Retro Biosciences to create GPT-4b, a specialized AI model focused on protein interactions and bioengineering. GPT-4b has demonstrated the ability to enhance Yamanaka factors, proteins crucial for cellular reprogramming and stem cell treatments. It has improved their effectiveness by up to 50 times, paving the way for advancements in regenerative medicine.

Thanks for reading! We’d love your feedback to make this newsletter even better. Our goal is to share content that’s valuable and relevant to you. Help us fine-tune future editions by sharing your thoughts on this week’s email, it’ll only take a moment!