AudioStack is building brand connection at the speed of sound

In a world congested with content and constant appeals for our attention, what words are spoken and what words are truly heard have never been more important.

Audio content that is personalized, localized, and shared at the right moment can be one of the most creative tools that marketing departments have at their disposal. But traditional audio production is both expensive and difficult to scale, which has led to the medium being left largely underutilized—until now.

Together with Amazon Web Services (AWS), AudioStack is democratizing audio production and helping brands tell their story with generative artificial intelligence (AI). The company is using Amazon Bedrock to accelerate and significantly reduce the cost of audio production workflows, helping businesses generate tailored and targeted audio content.

Breaking the sound barrier

AudioStack was founded in 2019 by an eclectic mix of machine learning (ML), audio engineering, and commercial experts. Today, the company is recognized as a true leader in end-to-end, enterprise-grade AI audio production. “AudioStack accelerates the production of audio for brands and agencies across the world,” says Peadar Coyle, Co-Founder and CTO. “Our mission is to democratize the creation of audio, and we feel a tremendous responsibility to make creativity faster and more inclusive.”

“Hearing is one the first senses that we develop in the womb. It’s why we are conditioned to hear certain sounds and feel safe or react in certain ways.” says Silke Zetzsche, Commercial Director. “As an advertising medium, audio has incredible creative potential, but it is often underutilized.”

“When it comes to traditional audio production, we always talk about the triangle of time, scope, and cost. For example, creating a single asset might be quicker, it might be cheaper, but it won’t be able to scale,” says Zetzsche. “AI audio allows us to deliver in all three of those areas. It allows audio assets to be scaled so that you can deliver millions of versions at a lower cost and a faster speed.”

AudioStack’s proprietary technology combines media creation capabilities such as AI script generation, text-to-speech, speech-to-speech, generative music, and dynamic versioning. Businesses rely on these capabilities to build cost and time efficient audio production workflows at scale.

For example, AudioStack’s SonicSell workflow enables users to generate ads based on simple text prompts. SonicSell performs sentiment analysis on the prompt and based on the results of that analysis it selects an appropriate voice and other audio to fit the ad. “You can have a fully produced audio asset in less than 60 seconds,” says Coyle.

“We are about 100 times faster than existing workflows and about 80 percent cheaper,” says Coyle. “This helps enterprises do things that they wouldn’t have been able to do before, like running localized campaigns that would have been too cost or time prohibitive in the past.”

Helping brands find their voice

AudioStack is helping businesses across multiple industries to speed up audio content production. “We’ve been working with a large media site and helped them to produce podcasts every day, on time, and on budget. This enables them to accelerate their time to value and customer reach,” says Coyle.

One of the main areas of focus for AudioStack is personalization and localization. “We have over 60 languages, and over 1300 voices and counting. We have use cases across all continents except Antarctica. A lot of our work focuses on inclusivity. For example, we’ve had some work with indigenous languages throughout the world and that’s been very exciting.”

“We are working with a UK-based retailer who wanted to be an innovator in the audio space and utilize AI as part of their production,” says Zetzsche. “They are now able to differentiate creative output between stores in different regions, to automate, and create content around multiple products in a single creative asset.”

Success stories like these have helped AudioStack build its reputation and stand out from the competition. “We have two main differentiators,” says Coyle. “One is that we’ve been enterprise focused from the start. The other is the number of partnerships that we’ve built, such as the partnership with AWS.”

Pitch-perfect collaboration

The world of generative AI is always changing. To be successful in the space, businesses need access to cutting-edge technology and expertise. AudioStack found both while working with AWS. “What initially attracted us to AWS was the security and scalability, as well as the vast ecosystem of partners and engineers,” says Coyle. “I’ve been using AWS for my whole career as an engineer, and when we were looking at cloud providers, we decided to go with them because they had the biggest ecosystem, the biggest community—we felt like we were in good hands from day one.”

“One of the best things that happened to us in the early days was getting into the AWS Activate program,” says Coyle. The AWS Activate program is designed to help startups get up and running on AWS as quickly as possible, and to provide support to early-stage disruptors at a time when funds can be scarce.

Participants in the program gain access to free AWS credits that can be spent on AWS services and expert support. “The AWS Activate program enabled us to experiment with new technology with little risk,” says Coyle. “We also got to work with solution architects to make sure that our architecture was scalable, robust, and cost efficient.”

Audio excellence powered by high-performance AI

AudioStack is using generative AI models available on Amazon Bedrock to power its audio production workflows. Amazon Bedrock provides access to high-performance foundation models (FMs) through a single API. This helps businesses like AudioStack securely experiment with and privately customize models from leading providers. Since Amazon Bedrock is serverless, there’s no need to manage any infrastructure, and it’s easy to securely integrate and deploy generative AI capabilities using AWS services.

AudioStack is employing techniques such as Retrieval Augmented Generation (RAG) to improve and tailor the outputs of its models. “We leverage Amazon Bedrock for RAG pipelines to produce more differentiated content much faster,” says Coyle. “We’re able to pull in things like brand guidelines, which enables our customers to deliver better experiences and remain brand safe.”

Seeing gains on go-to-market

Beyond technology, AudioStack is also working with AWS to build credibility, reach new customers, and unlock new go-to-market opportunities. The company has recently joined the AWS Partner Network (APN), a global community of validated AWS Partners committed to reaching millions of new customers, helping them solve technical challenges, and delivering more value with secure, high-quality software.

Before becoming an AWS Partner, businesses must complete the AWS Foundational Technical Review (FTR). The AWS FTR is a self-service review that helps businesses identify and remediate potential risks in their software following AWS Well-Architected Best Practices.

Once a business has completed the AWS FTR, they’re able to list their offerings in the AWS Partner Solution Finder and obtain a “Qualified Software” badge. This helps to immediately build credibility and confidence in customers browsing AWS Partner solutions. “Customers value trust a lot,” says Coyle. “In recent years, we’ve become SOC 2 Type 2 compliant and that, in combination with the AWS FTR, has helped us improve our security posture.”

By becoming an AWS Partner, AudioStack offerings are now listed on the AWS Marketplace. “We’ve been part of the AWS Marketplace for over six months,” says Coyle. “It has helped us gain credibility, gain customer trust, and enhance our offering in collaboration with AWS go-to-market teams.” Zetzsche adds: “The AWS Marketplace allows us to work together with AWS to develop our go-to-market strategy and uncover new business opportunities.”

Good vibrations

Going forward, AudioStack is continuing to raise the bar in AI audio production and equip enterprises with a powerful new line of connection with their customers. “We can unlock creativity for any of our clients by utilizing AI as part of the audio production process, specifically to tap into the personalization, automation, and scalability of creative assets,” says Zetzsche.

“Being able to adjust something on the fly at any point, or to tap into social movements contextually is a fantastic opportunity for brands and advertisers to increase relevance,” says Zetzsche. “It enables brands and marketeers to build more one-on-one conversations with potential and consumers, rather than one-to-many creative experiences.”

Together with AWS, the company is now turning its attention towards growth. “I’d recommend AWS to startups because of its vast ecosystem which has helped us access investors, customers, partners, and events that ultimately helped us get to market faster,” says Coyle. “With support from AWS, we’re looking at expanding into new markets internationally.” Zetzsche adds: “We’re very excited about growth, both from a commercial perspective but also from a regional perspective, as we begin working with more clients and brands across the globe.”