Audioburst After 4 Years

The post was originally published in Russian on Startup of the Day. Alexander kindly agreed to republish what we think is of great value to our readers.

Thousands of radio stations and podcasters talk unimaginable amounts of hours of smart thoughts every day. Some of the audience listen to the broadcast live, and some of them – within a couple of hours afterward, then the recording lands in the archive, and nobody will ever download it. Written articles live much longer thanks to search engines, Google will take audiences there even after ten years, it preserves the wisdom for ages to come.

Audioburst, the startup of the day, tries to restore justice. It created an audio search engine: it translates voice into written text, indexes the transcripts of broadcasts, looks for the fragment relevant to the user’s query, and plays the audio from the desired place. Content creators get free traffic and willingly help the startup to collect data. The trouble came from the clients’ side – nobody looks for anything in the search engine. The desire for an audio response to a precise question is very particular, there are better ways to look for new podcasts to subscribe to. It so happened that nobody wanted Audioburst.

A pivot was necessary. If the startup itself couldn’t bring the audience, let partners look for it.  The old project is still there, but it’s running in the background, now Audioburst focuses on API. It distributes an interface for audio search and an up-to-date list of trending recordings all over the world. Of course, they both get filtered, adjusted to the taste of an exact user, and then there’s a long list of features designed by the ‘because we can’ principle.

Most actual implementations look weak, but two of them seem viable. The startup developed up-to-date news for voice assistants. ‘Tell me, Alexa, what goes on in the world?..’ Most probably, half of the American media have similar tools, but an aggregator can exist, too. The second interesting scenario is a local search inside the radio station’s archive, an apparently useful thing, and the client is ready to pay for it. At the end of the day, it will probably become the main product, after all, we’ll see.

Audioburst spends a lot of money, it already burned USD 11M, and brought in another 3M one month ago.

—

This is a rerun from 2018. The startup brought in another USD 10M in 2019 – from the car manufacturer Hyundai at that, which should be a good sign, but there hasn’t been any good news ever since. And the fact that an ‘NFT Marketplace’ appeared on the website, offering radio stations to sell bits of their broadcasts, is an apparent sign of the team’s utter desperation.