/

Stable Audio 2.0

by Stability AI
  • What it is:Stable Audio 2.0 is a text-to-audio and audio-to-audio AI model by Stability AI that generates high-quality instrumental music tracks up to three minutes long at 44.1 kHz stereo with coherent structures like intros and outros.
  • Best for:Independent musicians and producers, Content creators (YouTubers, podcasters, video producers), Game developers and interactive media studios
  • Pricing:Free tier available, paid plans from varies
  • Rating:78/100Good
  • Expert's conclusion:Stable Audio 2.0 is ideal for musicians, producers & content creators who view AI as a collaborative tool for composing music & designing sounds much faster; provides assurance that training data is derived from licensed sources.
Visit website
Reviewed byMaxim Manylov·Web3 Engineer & Serial Founder

What Are Stable Audio 2.0's Key Business Metrics?

📊
3 minutes
Max Track Length
📊
44.1 kHz stereo
Audio Quality
📊
800,000+ audio files
Training Dataset Size
📊
April 2024
Release Date
Regulated By
Audible Magic(Copyright compliance)

How Credible and Trustworthy Is Stable Audio 2.0?

78/100
Good

This is an established company that has developed strong technical capabilities and provides copyright protections but because they are a private company there is very little publicly available information regarding their financials and recent company-wide changes have caused slight instability.

BREAKDOWN
Product Maturity75/100
Company Stability70/100
Security & Compliance85/100
User Reviews65/100
Transparency80/100
Support Quality70/100
TRUST SIGNALS
Licensed AudioSparx dataset with opt-outAudible Magic copyright protection44.1kHz professional audio qualityOpen technical architecture details

What Are the Key Features of Stable Audio 2.0?

✨
Text-to-Music Generation
The system produces three minute long instrumental tracks based on a user’s input for music genre, mood, instrumentation, and structural format.
✨
Audio-to-Audio Transformation
Users can upload a variety of audio clips such as voice or drum patterns, which can then be converted into different instrument types or styles based on the users text-based prompt.
✨
Structured Song Composition
The system will produce a full-length instrumental track with a clear introduction, development section, verse/chorus section, and conclusion.
✨
High-Fidelity Output
The produced files are 44.1 kHz stereo and can be used in a variety of capacities in the music industry including music production, creating stems, creating sound effects, and creating backing tracks.
✨
Copyright Protection
In order to protect the rights of the creators of the original audio samples that users upload, Audible Magic scans each uploaded sample to ensure that none of the sample contains any copyrighted material. Audible Magic uses the AudioSparx dataset which was specifically created for use in developing and testing music creation systems.
✨
Style Transfer & Variations
When producing music, the system will generate music that fits within an existing style of music and also create a variety of different versions of the same track so that users can experiment with different styles and options.
✨
Diffusion Transformer Architecture
The system utilizes a DiT model which is comprised of an autoencoder that compresses data in order to allow the model to process large amounts of data in order to generate music with a coherent musical structure.

What Are the Best Use Cases for Stable Audio 2.0?

Music Producers
One of the main benefits of the system is that it allows producers and composers to generate a wide variety of high-quality instrumental tracks quickly and efficiently from text-based descriptions in order to accelerate the production workflow.
Content Creators
Another benefit of the system is that it allows users to create royalty-free music and sound effects that can be used in video production, podcasting, and social media marketing without having to pay licensing fees or deal with delays associated with production.
Sound Designers
The system can take a variety of vocal recordings or simple audio samples and convert them into realistic instrument sounds and high-quality sound effects for game developers and filmmakers.
Songwriters & Composers
One of the ways that the system can help facilitate the production process is by allowing producers to rapidly prototype song structures, arrangements, and harmonic progressions in order to test out new ideas prior to actually recording and producing a full-length version of a track.
NOT FORVocal-Focused Artists
Because the system is only capable of generating instrumentals and cannot create vocals or sing, the system is limited in terms of its overall value to users who need to record vocals or create songs that include singing.
NOT FORReal-Time Live Performance
Due to the time required to generate the music, the system is not suitable for use in live applications or real-time music performance.

How Much Does Stable Audio 2.0 Cost and What Plans Are Available?

Pricing information with service tiers, costs, and details
☐Service$CostℹDetails🔗Source
Web AccessFreeGenerate tracks directly on stableaudio.com with text and audio-to-audio capabilitiesStability AI announcements
Stable Audio APIProgrammatic access for developers and production applications—
Web AccessFree
Generate tracks directly on stableaudio.com with text and audio-to-audio capabilities
Stability AI announcements
Stable Audio API
Programmatic access for developers and production applications

How Does Stable Audio 2.0 Compare to Competitors?

FeatureStable Audio 2.0MusicGen (Meta)SunoUdio
Max Track Length3 minutes30 seconds4 minutes4 minutes
Audio Quality44.1kHz stereo32kHz mono44.1kHz44.1kHz
Audio-to-AudioYesLimited melody modeNoNo
Musical StructureIntro/Dev/OutroBasicFull songsFull songs
Copyright ProtectionAudible Magic scanningResearch datasetLicensed dataLicensed data
Instrumentals OnlyYesYesVocals availableVocals available
API Access—Open sourceYesYes
Free TierYes (web)YesYesYes
Training DataLicensed AudioSparxResearch datasetProprietaryProprietary
Style TransferYesPartialLimitedLimited
Max Track Length
Stable Audio 2.03 minutes
MusicGen (Meta)30 seconds
Suno4 minutes
Udio4 minutes
Audio Quality
Stable Audio 2.044.1kHz stereo
MusicGen (Meta)32kHz mono
Suno44.1kHz
Udio44.1kHz
Audio-to-Audio
Stable Audio 2.0Yes
MusicGen (Meta)Limited melody mode
SunoNo
UdioNo
Musical Structure
Stable Audio 2.0Intro/Dev/Outro
MusicGen (Meta)Basic
SunoFull songs
UdioFull songs
Copyright Protection
Stable Audio 2.0Audible Magic scanning
MusicGen (Meta)Research dataset
SunoLicensed data
UdioLicensed data
Instrumentals Only
Stable Audio 2.0Yes
MusicGen (Meta)Yes
SunoVocals available
UdioVocals available
API Access
Stable Audio 2.0—
MusicGen (Meta)Open source
SunoYes
UdioYes
Free Tier
Stable Audio 2.0Yes (web)
MusicGen (Meta)Yes
SunoYes
UdioYes
Training Data
Stable Audio 2.0Licensed AudioSparx
MusicGen (Meta)Research dataset
SunoProprietary
UdioProprietary
Style Transfer
Stable Audio 2.0Yes
MusicGen (Meta)Partial
SunoLimited
UdioLimited

How Does Stable Audio 2.0 Compare to Competitors?

vs MusicGen by Meta

All can create music from one type of input to another (audio → audio), however Stable Audio 2.0 has the ability to generate 3 minute pieces with structured composition (i.e., intro, build-up, conclusion) whereas Music Gen creates much shorter pieces. Stable Audio 2.0 also features style transfer and utilizes a diffusion transformer as an architecture to improve how it handles long sequences of audio. Music Gen still maintains performance capabilities in generating music in real time.

If you want to create complete songs that follow a traditional format choose Stable Audio 2.0, if you want to quickly experiment with music ideas and have the need to produce music in real time choose Music Gen.

vs AIVA and Amper

Stable Audio 2.0 excels in terms of the length of generated pieces (3 minutes vs. 1 – 2 minutes), and by offering users more ways to interactively create music (both using text prompts or transforming one piece of audio into another). AIVA & Amper are two tools that offer their users more music theory based workflow options, and thus more creative options for composers. Stable Audio 2.0 makes creating music more accessible for non-musicians through the use of natural language prompts.

If you want to create music using AI technology in a way that is accessible to everyone, choose Stable Audio 2.0. If you are looking to compose music that incorporates elements of music theory, then choose either AIVA or Amper.

vs Soundraw

While both tools provide some form of interactive creativity, Soundraw allows its users to adjust settings and customize their tool through sliders and presets, where as Stable Audio 2.0 provides these same customizing options through natural language prompts and audio transformations. In addition to providing the same level of customization options, Stable Audio 2.0 produces longer pieces of music than Soundraw, and provides users with more creative freedom when experimenting with sounds. On the other hand, Soundraw provides its users with more granular control over their creative process, however this comes at the cost of the automation offered by Stable Audio 2.0.

If you prefer to utilize AI technology to create music without having to manually intervene too much, choose Stable Audio 2.0. If you prefer to manually customize your musical creations in great detail, choose Soundraw.

vs OpenAI's Jukebox

Unlike Jukebox, which was created primarily for research purposes and produced music of lesser quality, Stable Audio 2.0 is designed to be used practically and is maintained regularly, and is provided free, yet also commercially viable. In addition to being actively maintained, Stable Audio 2.0 also protects copyrights through the use of Audible Magic, and is trained on licensed data.

Practicality wise, Stable Audio 2.0 is your best option for creating music using AI technology; Jukebox is mostly obsolete and is no longer supported.

What are the strengths and limitations of Stable Audio 2.0?

Pros

  • Can generate complete pieces of music up to 3 minutes in length (complete songs with intro, build-up, and conclusion)
  • Has dual generation modes (generate music based on text prompts and transform one piece of audio into another); This provides users with the greatest amount of creative flexibility for creating music in whatever way they see fit.
  • Ethical licensed training data for use from AudioSparx allowing artists to opt out — respects creator rights.
  • The training data is free to test for use at the Stable Audio website — there are no barriers (paywalls) to entry for testing and using this tool for your own projects.
  • Diffusion Transformer Architecture — enables the model to better process long sequences that create coherent musical structures.
  • Style Transfer Capability — allows users to customize their generated audio or upload files to match the theme of the user's project.
  • High Quality Audio at 44.1 kHz Stereo — created to meet professional standards to allow for high-quality productions.
  • Audible Magic Content Recognition — provides built-in copyright compliance to prevent users from uploading copyrighted material and infringing on those copyrights.
  • Separation of Instrument Components and Generation of Backing Tracks — can provide the separated components of instrumental parts useful for music production.

Cons

  • Timbre Transfer Limitations for Audio-to-Audio Transfers — while providing primarily timbre transfer, does not yet allow for the creation of full multi-instrumental arrangements including drums and bass.
  • Must Upload Copyright-Free Material — The Terms of Use require that all material uploaded to the tool to be used in audio-to-audio transfers be copyright-free, thus limiting the source of samples.
  • Feature Set Limitations Compared to Competitors for Instrument Separation — unlike some other specialized music production tools, the tool does not allow for isolating specific instruments from generated audio tracks.
  • Music Theory Controls Not Available — lacks granular control over musical harmonic progressions when compared to music theory focused competitors.
  • Availability of the Tool Through the API Pending — the tool will initially only be available through the website until later when it becomes available via API, creating potential workflow issues for developers.
  • Research Paper Has Yet to Be Published — due to the lack of completion of the technical details regarding the tool, the ability to reproduce the tool and subject it to peer review is limited.
  • Degradation of Quality for Complex Prompts — ambiguity within the natural language used to prompt the tool can lead to variable results on complex musical requests.
  • Three-Minute Time Limit — limits the ability to create longer-form content such as extended DJ mixes or orchestral suites.

Who Is Stable Audio 2.0 Best For?

Best For

  • Independent musicians and producers — Rapid Composition and Experimentation Enabled by Easy Access to the Tool and Natural Language Prompts — enables users to rapidly compose and experiment with music ideas without the need for expensive digital audio workstations or licensing fees.
  • Content creators (YouTubers, podcasters, video producers) — Creates royalty-free background music, intro and outro tracks, as well as other sound effects suitable for use in multimedia productions at a high level of quality.
  • Game developers and interactive media studios — The capability to generate both audio-to-audio content, and to create various sound effects allows for the rapid generation of large amounts of media assets; stems can also be used to implement media assets in a variety of game engines.
  • Advertising and marketing agencies — Fast turn-around on custom background tracks and other custom audio assets without having to go through extensive licensing or composer hire processes.
  • Music educators and students — As a free resource, it provides an opportunity to learn how to compose music, understand production workflows, and experience AI capabilities without any financial restrictions.
  • Experimental musicians and sound designers — Its unique ability to generate audio-to-audio content and utilize style transfer enables users to explore new and unique sonic combinations and to create hybrid human-AI generated content.

Not Suitable For

  • Professional film composers requiring custom orchestral work — The audio tracks generated by this tool do not provide the same level of nuanced emotional control and orchestration that a human composer would provide; consider hiring a professional composer or utilizing AIVA which has the option to utilize full music theory controls.
  • Music professionals requiring stems with specific instrument isolation — Current audio-to-audio functionality does not ensure that there will be multiple instrument arrangements with completely separate drum and bass tracks; consider utilizing a specialized stem separation tool.
  • Musicians prioritizing full creative control over AI assistance — The unpredictable nature of natural language prompts compared to traditional digital audio workstations (DAWs) may require users to utilize Logic Pro, Ableton, or Cubase for deterministic composition.
  • Enterprises needing custom integrations immediately — As of now, the API is only available via stable release and is currently only accessible via website interface; wait for API launch or use MusicGen's API infrastructure that is more mature.
  • Users creating content over 3 minutes (podcasts, long-form videos) — The hard 3-minute time limit imposed by Stable Audio 2.0 limits its usability for creating extended-length audio projects; utilize traditional music libraries or composing tools instead.

Are There Usage Limits or Geographic Restrictions for Stable Audio 2.0?

Maximum Track Length
3 minutes per generation at 44.1 kHz stereo
Audio Upload Restrictions
Must be free of copyrighted material; advanced content recognition scans for compliance
API Availability
Currently web-only; Stable Audio API coming soon
Audio Format
Stereo output at 44.1 kHz; no surround sound or higher sample rates
Data Source
Trained on AudioSparx licensed library of 800,000+ files; artists could opt-out during training
Geographic Availability
Available globally with no mentioned regional restrictions
Commercial Use Rights
Free tier allows personal/commercial use of generated audio with proper attribution
Terms of Service Compliance
No copyright infringement on uploads; automated detection via Audible Magic technology prevents violations

What APIs and Integrations Does Stable Audio 2.0 Support?

API Status
Stable Audio API coming soon; currently web-based interface only
Current Integration Method
Web platform at stableaudio.com; direct API integration not yet available
Planned SDK Support
SDKs expected with API launch; language availability not yet specified
Authentication
Web-based user accounts; API authentication method to be announced
Webhooks
Not mentioned for current version; likely available upon API release
Documentation
User guide available at stableaudio.com/user-guide; API documentation pending
Use Cases
Music generation from natural language, audio-to-audio transformation, sound effect creation, style transfer, stem generation
Supported Input Formats
Natural language text prompts; audio file uploads for audio-to-audio generation (format details not specified)
Output Formats
44.1 kHz stereo audio files; specific download formats not detailed

What Are Common Questions About Stable Audio 2.0?

Stable Audio 2.0 has the capability to generate a complete musical track up to 3-minutes long at 44.1 kHz stereo, and includes a structured composition format that includes an introductory section, a developmental section, and a concluding section.

The text-to-audio feature utilizes natural language prompts to create music from scratch; the audio-to-audio feature utilizes natural language prompts to modify the user-uploaded audio sample, and thereby perform style transfer and customize the sound of the original audio.

Yes, Stable Audio 2.0 is available for free on the Stable Audio website; a paid API version will be available in the near future, but no pricing information has been released.

When users upload their own music, the application's Audible Magic content recognition software is used to scan for possible copyright infringement violations so that the users do not infringe on someone else’s copyrights. Artists whose music has been used by AudioSparx are given an opportunity to opt out of having their work included in the model.

Yes. The free version of this product may be used by individuals or companies for personal and commercial purposes. Please refer to the current Terms of Service to confirm all rights.

Both products support generating audio to audio, but Stable Audio 2.0 generates full length songs (up to 3 min.) and includes traditional song structure, while MusicGen produces smaller pieces of output.

Stable Audio 2.0 generates audio at 44.1 kHz Stereo which is broadcast quality and suitable for professional music production and distribution.

The Stable Audio API is expected to become available “soon” based on company announcements; however no specific release date has been announced. Please refer to stableaudio.com for further updates.

Yes. Both versions support creating melodies and backing tracks and/or individual stems for the flexibility of producing music.

Version 2.0 uses a very compact autoencoder for representing the audio and a diffusion transformer (similar to Stable Diffusion 3) instead of U-Net, allowing it to generate full length tracks of up to three minutes with a coherent musical structure over long sequence lengths.

Is Stable Audio 2.0 Worth It?

Stable Audio 2.0 represents a major improvement in AI music creation and delivers professional quality full length music tracks with coherent musical structures. In addition to its text-to-audio functionality, Stable Audio also supports audio-to-audio capabilities and includes a commitment to compensating creators through licensed training data, making it a valuable resource for musicians and content producers. While Stable Audio 2.0 offers many advantages as a tool for music and sound design, it is still limited to those areas of audio production and does not offer a solution for generalized audio.

Recommended For

  • Independent musicians and composers who wish to speed up their creative process
  • For content creators (video, podcasts) looking for background music & sound effects
  • For music producers wanting to use AI as a collaborative compositional resource
  • For game developers & film studios to produce customized instrumental scores
  • For audio designers producing many types of sound effect libraries
  • For hobbyists & professionals w/ modest to mid-range budget

!
Use With Caution

  • Users creating vocals — this tool produces only instrumental music
  • For creators needing complete creative control over their project — generated AI will require iteration
  • Extremely niche / specialized genres are required — can be time-consuming with extensive prompting
  • Needing verification that all audio samples are copyright compliant — must verify

Not Recommended For

  • Teams generating voice/music — this is an entirely different use-case
  • For budget-restricted users unable to fund subscriptions — need access to the platform
  • Needing 100% guarantee of copyright free generated outputs — must audit compliance of training data
  • Users expecting immediate production-ready results — no additional refinement needed
Expert's Conclusion

Stable Audio 2.0 is ideal for musicians, producers & content creators who view AI as a collaborative tool for composing music & designing sounds much faster; provides assurance that training data is derived from licensed sources.

Best For
Independent musicians and composers who wish to speed up their creative processFor content creators (video, podcasts) looking for background music & sound effectsFor music producers wanting to use AI as a collaborative compositional resource

What do expert reviews and research say about Stable Audio 2.0?

Key Findings

Stable Audio 2.0 produces high-quality instrumental music (up to 3 minutes @ 44.1 kHz stereo) with structured composition elements (intros, dev. sec., outros). Stable Audio 2.0 has both text-to-audio and audio-to-audio generation functionality — enabling users to transform uploaded samples via natural language prompts. Stability AI was solely responsible for training the model on a licensed dataset (from AudioSparx), provided creator opt-out options & ensured fair compensation to artists — addresses copyright concerns associated with AI music generation.

Data Quality

Excellent — comprehensive information from official Stability AI announcements, company blog posts, and multiple tech publications (Tom's Guide, Music Business Worldwide). Technical specifications and feature sets are consistently documented across sources. Product availability and pricing confirmed through official channels.

Risk Factors

!
Produces instrumental music only — does not support vocal music generation
!
Requires natural language prompting — the quality of output will depend on the specificity of the prompt
!
The Stable Audio 2.0 Platform: An Overview
!
Stable Audio 2.0 is a free and paid music generation tool that is part of the Stable Audio platform that is owned by the company Stability AI. The Stable Audio platform allows users to generate music using a number of different tools including music generation, theme modification and effects generation. The platform has been recognized as one of TIME Magazine's Best Inventions of 2023 and provides users with the opportunity to create high quality music from scratch or edit existing music tracks.
!
Features of Stable Audio 2.0
Last updated: February 2026

What Additional Information Is Available for Stable Audio 2.0?

Awards & Recognition

Stable Audio 2.0 offers users a wide range of features that allow them to generate music and edit tracks. Music Generation: Users can generate entire music tracks using the music generation tool included in the platform. Theme Modification: Users are able to modify the theme of their generated music to fit the mood and tone they want to convey. Effects Generation: Users are also able to add effects such as sound effects and keyboard taps to their generated music. Variations Generation: Users are able to generate multiple variations of their music using the variations generator included in the platform. Stems Generation: Users are able to generate individual stems for each element of the generated music, allowing for further editing. Backing Tracks: Users are able to generate backing tracks for their generated music, which can then be used as the foundation for further editing and composition.

Technical Architecture

Availability of Stable Audio 2.0

Key Features

Stable Audio 2.0 is currently available to users at no cost on the Stable Audio website. A free tier will remain available after the API is released and will remove the financial barrier to entry for users who wish to try out the platform prior to purchasing a paid version of the software.

Accessibility

Multi-Modal Creative Workflow

Integration with Stable Assistant

In addition to generating music, Stable Audio 2.0 is integrated into the larger Stable Assistant platform. This means that users are able to access image generation and editing tools through the same user interface as the music generation tools, providing users with the ability to perform multi-modal creative tasks. For example, users can generate images based on music they have created, or create music inspired by images.

Creator Protection

Training Data Used for Model Development The Stable Audio 2.0 model was developed using a large dataset of music from the AudioSparx library. The use of this data required the developers to obtain permission from the copyright holders of the songs included in the library and to provide these copyright holders with an option to opt-out of having their work included in the model. Developers were also required to provide a mechanism for paying the copyright holders of the works included in the model. The developers also required that all audio uploaded to the platform be copyright-free and that they use a content recognition system to ensure that users complied with this requirement. :

What Are the Best Alternatives to Stable Audio 2.0?

  • •
    OpenAI Jukebox / Spotify's Jukebox: Experimental AI music composition with vocal elements. Good for experimenting with different styles, however the focus is more on the creativity behind the music rather than the professionalism of how it was produced. Best for academic purposes/researchers, individuals who want to create music with a vocalist first and foremost.
  • •
    Google MusicLM: AI music composition by Google that produces very good quality music from text descriptions. The main usage of this tool is for research. This tool has similar quality as Stable Audio but is much harder to obtain than Stable Audio for everyday use. Best for an organization with a partnership to conduct research or an organization that uses the Google Ecosystem.
  • •
    AIVA (Artificial Intelligence Virtual Artist): A complete AI composition platform for creating the original score for films and games. This platform allows for the maximum amount of customizations available and also includes many other professional tools to assist you in your work. Best for film, game and media companies looking for a specialized solution for their scoring needs.
  • •
    Amper Music: AI music composition designed for the use of Content Creators & Brands. Simple Interface and preset styles and moods. Not as much creative control as Stable Audio but much faster for casual use. Best for people without musical experience and brands requiring quick, royalty free background music.
  • •
    Landr Studio: An AI mastering/mixing platform that offers music composition. Combines AI composition with AI mastering/mixing. More expensive than Stable Audio, but gives added value post-production. Best for Producers who would like to have a single platform to generate music and professionally finish the product.
  • •
    Splice Sound (with AI features): Sample Library/Collaboration Platform that is introducing AI Composition Features. Ideal for Loop Based/Sample Driven Music Production. Best for Producers that are already heavily invested into the Splice Ecosystem and would like AI assistance for creating/altering samples/remixes.

What Is Stable Audio 2.0's Model Overview?

Developer
Stability AI
Version
Stable Audio 2.0
Release Date
April 3, 2024
Architecture
Latent Diffusion with Diffusion Transformer (DiT) and compressed autoencoder
Open Source
No
Status
Generally Available

How Does Stable Audio 2.0's Model Versions Compare?

VersionRelease DateKey Improvements
1.0September 2023Initial release, high-quality 44.1kHz music up to 90s
2.0April 2024Up to 3min tracks, audio-to-audio, style transfer, improved structure

What Is Stable Audio 2.0's Audio Generation Specs?

Max Duration
3 minutes
Sample Rate
44.1 kHz
Channels
Stereo

What Generation Modes Does Stable Audio 2.0 Offer?

Text-to-Audio

Create Full Tracks & Sound Effects from Natural Language Prompts

Audio-to-Audio

Modify Uploaded Audio Samples Using Natural Language Prompts

Sound Effects Creation

Create SFX from Text Descriptions

Style Transfer

Change the style of Generated or Uploaded Audio (Based on Style Description)

What Music Capabilities Does Stable Audio 2.0 Offer?

Full-Length Tracks

Tracks should be no longer than 3 minutes with an intro, a body (or development), and an outro.

Genre Support

Audio types can be very varied from orchestral music to ambient lo-fi tracks to funk and even to drum solos.

Melodies & Backing Tracks

The software generates melodies, backing tracks and separate tracks for each instrument or voice as required by the customer.

Stereo Sound Effects

Software will apply spatial audio effects to all tracks that are generated.

How Does Stable Audio 2.0's Benchmark Scores Compare?

BenchmarkScoreRankNotes
Audio Quality MetricsState-of-the-art#1According to Stability AI evaluation
Prompt AlignmentState-of-the-art#1Subjective human tests
Full-Length CoherenceLeading3min structured compositions

What Is Stable Audio 2.0's Access Licensing?

Open Source
No
License
Proprietary with Terms of Service
GPU Requirements
Cloud-based
Platforms
stableaudio.com (free access), Stable Audio API (coming soon)

How Does Stable Audio 2.0's Generation Pricing Compare?

TierCostDurationQualityNotes
Free$0Up to 3minStandardAvailable now on stableaudio.com
APIUp to 3minHigh

What Creative Tools Does Stable Audio 2.0 Offer?

Audio-to-Audio Transformation

The customer may upload their own samples and use these in conjunction with text prompts to create new sounds.

Style Transfer

The customer is able to tell the software what style they want the generated/uploaded audio to be matched to.

Full Track Generation

Tracks generated will have structured compositions with an intro, a body (or development), and an outro.

Variations

Customers may generate multiple different versions of audio created by the software.

What Is Stable Audio 2.0's Content Safety Status?

Copyright FilterAudible Magic content recognition for uploads
Licensed Training DataTrained exclusively on AudioSparx library
Opt-Out SupportArtists can opt-out of training data
Upload RestrictionsNo copyrighted material allowed per ToS
WatermarkingNot mentioned

Expert Reviews

📝

No reviews yet

Be the first to review Stable Audio 2.0!

Write a Review

Similar Products

Soundraw
soundraw.io
Udio V2
udio.com
Riffusion
riffusion.com