A while back Suno V3 killed it and stirred up the music scene.
On April 2, more than 200 prominent musicians signed an open letter calling for a halt to AI's assault on human creativity.
The open letter, released by the Artists Rights Coalition advocacy organization, asks technology companies to commit to not developing AI tools that undermine or replace human songwriters and artists.
The autographs have been signed by not only top stars like Billie Eilish, J Balvin and Nicki Minaj, but also Rock and Roll Hall of Famers like Stevie Wonder and REM.
Just one day after this open letter was sent, Stability AI released on Platform XAI MusictoolStable Audio 2.0.
Product Introduction
Stable Audio 2.0 is an advanced audio generation model developed by Stability AI.
The model is capable of generating high-quality, up to three-minute-long musical compositions based on textual cues or uploaded audio samples and supports a wide range of musical styles, such as rock, jazz, electronic, and hip-hop.
Its main features include:
1. High-quality music generation:Stable Audio 2.0 generates 44.1kHz high-fidelity musical compositions that are fully structured, including introductions, developments and endings, as well as stereo sound effects.
2. Audio to audio conversion:This model allows users to upload audio samples and transform these samples into different sounds using natural language cues.
3. Efficient generation speed:Compared to the previous version, Stable Audio 2.0 significantly improves the efficiency of music generation, averaging about 1 minute for a 3-minute piece of music.
4. large-scale dataset training:The model was trained using over 800,000 audio files and 19,500 hours of audio data to ensure that the music generated is rich in detail and realism.
5. Commercialization application support:Partnering with renowned music service provider AudioSparx, Stable Audio 2.0 generates commercially available music for video self-publishing users and commercial advertisement production.
6. Diverse output formats:The generated music supports multiple formats for download, including MP3, WAV and Video, to meet the needs of different users.
Experience it online for free:
Official News:
https://stability.ai/news/stable-audio-2-0
Cue word guide.
https://stableaudio.com/user-guide/prompt-structure
Playing Guide
Step one:Log in to the Stable Audio 2.0 website and register with your Google or other email address. Then click "Try now" to enter the operation page.
Step 2:Enter the cue words. Since cue words directly affect the quality of the generated music, an official guideline is given specifically: the more details the better; it is best to include elements such as genre, descriptive phrases, instrumentation, mood, and beat.
Examples include Cinematic, Soundtrack, Wild West, High Noon Shoot Out, Percussion, Whistles, Horses, Action Scene, SFX, Shaker, Guitar, Bass, Timpani, Strings. Tense, Climactic, Atmospheric, Moody (Film, Soundtrack, Wild West, High Noon Shoot Out, Percussion, Whistles, Horses, Action Scene, SFX, Shaker, Guitar, Bass, Timpani, Strings, Tense, Climactic, Atmospheric, Moody)
There is also an official guide to common music cue words.
If users have no idea at all, they can click on "Prompt Library", the system provides 18 types of pop, classic rock, quiet, drum solo, etc., users can choose according to their own preferences.
For example, I chose pop (pop style), the system automatically enters the prompt words: Machine, Bass, Lush Synthesizer Pads, Synthesizer Arp, Synth Bass, Vocal Sample Chops, Percussion, Honest, Heart- Felt, Melancholic, Vibe, Cool, Modern, Atmospheric, 115 BPM.
Step Three:Adjust the parameters. First select the model, the system defaults to the latest Stable Audio 2.0 version. Then select the duration of the generated music, the maximum length is not more than 3 minutes. Finally, click "Generate".
In addition to text-born audio, Stable Audio 2.0 also enables audio-born audio.
The user simply uploads a piece of audio and then enters the prompt words. For example, I uploaded a snippet of the song "If the Moon Hadn't Come" and asked for it to be adapted to a disco style:
Is it free?
Magic is needed.
Stable Audio 2.0 is open for free trial and offers 20 free credits per month for new users, consuming 2 credits for every 3-minute music generated.
Also, Stable Audio 2.0 offers three payment plans, all of which generate audio that can be used commercially:
Pro version: $11.99 per month and offers 500 credits per month.
Studio Edition: $29.99 per month, offering 1,350 credits per month.
Top version: $89.99 per month, offering 4,500 credits per month.
Summarize
fromusabilityStable Audio 2.0 provides a simple interface that allows users to generate audio in a variety of styles by simply typing in descriptive text prompts such as music style, instrument, mood, etc. or by uploading a piece of music.
frominnovativenessOn the upside, Stable Audio 2.0 is capable of generating up to three minutes of high-quality music, including structured compositions such as intros, developments, and outros, as well as stereo sound effects. Keep in mind that Suno can only generate up to two minutes of audio. Moreover, the music generated by Stable Audio 2.0 can be used commercially.
fromfunctionalOn top of that, Stable Audio 2.0 also supports style shifting, which allows you to seamlessly modify newly generated or uploaded audio during the generation process, however, compared to Suno, Stable Audio 2.0 generates music effects with noticeable noise and sometimes fighting instruments.
In addition, Suno can automatically generate lyrics, a feature not available in Stable Audio 2.0.
In short, Stable Audio 2.0 is not yet able to shake SunoAI's position as the "first brother" in the music industry.