Adobe Firefly Launches AI Audio Tools With Commercially Safe Licensing to Challenge Suno

Adobe Firefly Expands Into Audio With Three AI-Powered Audio Tools

Adobe announced Thursday that Adobe Firefly now includes three AI audio tools available to all users: Generate Music, Generate Speech, and Generate Sound Effects1

1

2

2

. The tools emerged from beta testing at last year's Adobe Max conference and now sit alongside Firefly's image, video, and design capabilities1

1

. Each tool runs on a different model: Generate Music uses the Firefly Music Model, Generate Speech operates on the Firefly Speech Model with ElevenLabs offered as an option, and Generate Sound Effects utilizes the Firefly Audio Model2

2

. Access requires a Firefly subscription, which may be included in existing Creative Cloud subscriptions or available standalone starting at $10 per month1

1

. AI audio generations count as generative credits within the subscription1

1

.

Generate Music Creates Universally Licensed Tracks for Video Production Workflows

Generate Music produces instrumental background tracks rather than complete songs, targeting professional creators who need commercially safe audio1

1

. Users can create music through a Mad Libs-style prompt format, selecting vibe, genre, and situation, such as "a dreamline song, with electronic, ambient style, for a video game"1

1

. The tool can also analyze uploaded videos and automatically generate four sample audio tracks up to 30 seconds long that match the video's mood1

1

3

3

. Users can adjust parameters including tempo, energy, and duration before generating options3

3

. All music created receives a universal license, meaning creators can use it across platforms without copyright infringement concerns or takedown risks1

1

2

2

.

Commercially Safe Licensing Becomes Adobe's Primary Selling Point Against Competitors

Adobe emphasizes "commercially safe" licensing throughout its announcement, positioning legal protection as the core differentiator from competitors like Suno2

2

. The timing follows a July ruling by a German court that found Suno broke copyright, marking the first such decision in Europe brought by collecting society GEMA2

2

. Adobe trains Firefly models exclusively on licensed and public domain material, never using customer work to improve services1

1

. A survey conducted with Berklee College of Music found that 43.2% of video creators, musicians, and marketers cited legal and copyright risks as the main barrier to using music, while 38.9% pointed to licensing costs3

3

. The same survey revealed that 32.7% already use AI-generated music as the final track in published work, and 100% of respondents use music in their videos3

3

. Jay LeBoeuf, Adobe's head of AI audio, stated the goal is building tools that address pain points in audio creation and editing processes1

1

.

Generate Speech Offers Emotion Tags and Multi-Language Support for Voiceovers

Generate Speech transforms scripts into audio files with several artificial voices across various genders and ages, supporting translation into over 20 languages1

1

5

5

. Users can add pronunciation guidance for difficult names or products through phonetic spelling1

1

. The tool includes "emotion tags" that direct the AI to apply different expressions, addressing the challenge of making voices sound less robotic1

1

. Adobe collected more emotive training material specifically for this purpose, with the design team knowing from the start that adjectives and verbs would control the output1

1

. Users choose from 45 speakers described as male, female, non-binary, young adult, middle-aged, or senior4

4

. The interface allows adding pauses and adjusting tone of voice through a context menu, though testing revealed the need to manually select passages for preview rather than using a simple play button4

4

.

Generate Sound Effects and Third-Party Models Expand Creative Workflows

Generate Sound Effects creates custom sounds matching the action, timing, and energy of content3

3

. Users can either describe effects through text prompts or upload reference recordings, including using human voice that the AI transforms into desired sounds like deepening a roar to sound like a dinosaur or monster1

1

. Adobe positions these tools as solutions to workflow disruption, where sound design typically sends creators to specialized tools and separate services before stitching everything together3

3

. Firefly now hosts systems from Google, ElevenLabs, Kling AI, Luma AI, OpenAI, and Runway alongside Adobe's own models2

2

. Adobe added Google's Gemini Omni Flash on Thursday, which accepts video, audio, and image inputs alongside text, allowing rough ideas to become storyboarded first cuts in the same window2

2

3

3

. The inclusion of ElevenLabs, which was in talks over a tender offer at a $22 billion valuation in July, signals Adobe's willingness to serve as the workspace rather than controlling every model, provided work happens within its window2

2

. However, Adobe's commercial safety guarantee applies specifically to Firefly models rather than extending across all third-party systems2

2

.