Italian company Musixmatch is well known for providing community-powered lyrics to major music streaming platforms including Spotify, Apple Music, YouTube Music, Amazon Music, and Tidal. It is now launching a new platform for podcasts that combines AI-generated transcription and community-verified editing.
Although there are millions of podcast shows and episodes available to listeners, Musixmatch says the search for podcasts is down. As a result, this suggests that many quality podcasts are not connected to potential fans. So he uses his experience in training AI models through speech and leverages his expertise in NLP (Natural Language Processing) to improve transcription, search, discovery and sharing of podcasts.
Musixmatch’s podcast platform automatically generates transcripts of some of the best podcast episodes on different topics and charts every day. It uses its core NLP model architecture, Umberto, to tag keywords such as places, people, and topics with Wikipedia IDs – alphanumeric identifiers tied to topics on Wikipedia. (For example, this link points to a Wikipedia ID related to TechCrunch.)
Because of this approach, people searching for these topics in any language will get accurate results.
The startup told TechCrunch that based on these IDs, it creates a topic chart called TopicRank that ranks podcasts based on factors like the number of mentions in an episode or the presenters’ expertise on the topic – improving search results for podcasts when users search for Related Topics.
Picture credits: Musixmatch
“With this classification, people can finally search for a particular keyword and find transcribed podcasts that match their query, ranked by relevance. Our search index returns a much more detailed and in-depth array of results than any other search service. listening that relies on standard RSS metadata and predefined genres and categories,” the company said.
When users search Musixmatch’s podcast platform, it displays snippets of transcripts where the searched phrase is mentioned. If they click on the result, the podcast will start playing right from the timestamp of the snippet that mentions the phrase. This is very handy when you need to listen to a few minutes of audio while searching for something.
Musixmatch has long relied on its community to make precise edits to lyrics, and now it’s asking those users to do the same with podcasts. The company’s new podcast portal also includes a tool called Podcast Studio, which allows publishers and podcast owners to fix AI-generated transcription, especially handy for things like people and brand names or cultural references.
If there is no transcript for a particular episode, the owner or community member can use Podcast Studio to generate one. Musixmatch says it takes the AI about five minutes to generate a transcript for an episode. Regular listeners can also vote for an episode for the transcript to be prioritized by the community.
Picture credits: Musixmatch
It is important to note that on Musixmatch’s platform, the AI-generated transcript will have tags such as “Speaker 1” and “Speaker 2”, while community-edited episodes will have tags with the names speakers – along with a “verified” label.
Human curated podcast transcript and tags Picture credits: Musixmatch
The company also makes it easier to share by displaying cards containing text snippets from the podcast with a shareable link. Additionally, he’s working on a feature called audiograms, which are small, shareable videos that include audio and scrolling text snippets from a podcast.
Picture credits: Musixmatch
Musixmatch does not want to keep all this data to itself. It allows podcast owners to export transcripts to their web feeds and apps. And since these texts are SEO optimized, it will make it easier for listeners to find.
Some Musixmatch partners he claims use his transcription tools include Financial Times’ “The Talent Show”; Red Bull’s “Beyond the Ordinary” and “Why I Run”; and all the production of Chroa Media.
Although Musixmatch’s podcast platform offers features for listeners, it doesn’t try to be just a podcast player. The startup says its competitors are companies that work in audio analysis, including apps that provide transcription services (like Podcastle).
“We believe that audio analytics (AI, semantics, etc.) will be a must in the near future, for many different use cases. We are in a unique position to provide this service for podcasts due to our AI-based content analytics technology, our engaged community, and our role within DSPs. [demand side platforms]which we already syndicate third-party content for,” Marco Paglia, the company’s chief product officer, told TechCrunch via email.
He added that one of the company’s goals is to become a verified transcription provider for other services, just like its lyrics offerings.
#Musixmatch #launches #podcast #platform #communitydriven #transcription