Facebook launches Collab, a mix-and-match app for making collaborative music videos

Facebook’s internal R&D group, NPE Team, is rolling out yet another new app today called Collab after having just launched a new group audio calling app, CatchUp, on Tuesday. With Collab, the focus has now returned to video, and specifically, the concept of watching, mixing, and matching original videos together, beginning with music.

In Collab, creators can either record their own musical arrangement or swipe to discover arrangements to build a composition, or a “collab.” While there are some elements of TikTok’s duets in this idea, the difference is that all videos posted to Collab can be mixed and matched with others. TikTok, meanwhile, allows creators to control who can duet with them.

In addition, Collab is only designed for making original music videos for the time being, which sets it apart from other video apps — including TikTok, Dubsmash, Triller, and more which have users creating content to the music from popular songs available via an in-app catalog.

Though focused on music, you don’t necessarily have to be a gifted musician to publish to Collab. You could participate by doing something simple — like banging on a child’s xylophone, beating a tambourine, pulling on a roll of tape, tapping a glass bottle, or even just tapping their foot. Musicians could then use that video alongside their own content to build their “collab.”

The collabs can only be up to 15 seconds in length, as this is not intended to be a professional music-making platform, but rather one that’s used for fun and experimentation.

Once users have created a collab, they can publish it for others to watch in the app’s feed or to further remix. However, the underlying music itself cannot be remixed — only the videos. The resulting collab can also be published to other social media platforms, including Instagram, Facebook Stories, and more.

There are a number of existing apps that allow users to collaborate with others on music, including by mixing sounds, making recordings, and arranging compositions. But these tend to be digital audio workstation (DAW) software programs, or at least those aimed at semi-professional to professional musicians. Spotify’s Soundtrap is one example. BandLab, Endless, Bandpass, Kompoz, are a few others. Vampr, meanwhile, helps musicians discover new collaborators. Collab, meanwhile, is more open to mainstream users — including those who play music for fun or are just fans of music in general.

Facebook says it’s been working on Collab for a few months, but hurried the launch in light of so many people being sheltered in place around the world due to the coronavirus pandemic.

“Digital spaces can connect us when we can’t be together in person, and Collab is a new way to create together,” a Facebook company spokesperson said about the launch.

However, the app itself is not ready to rapidly scale, which is why it’s being released today as an invite-only beta.

The company notes there’s still work that needs to be done to polish the app’s experience, but the team will be iterating on the product and responding to user feedback going forward. More people will be able to join Collab as invites roll out in batches. Access to the waitlist is here.

Collab is the latest in a series of releases from Facebook’s R&D group, NPE Team, which so far has launched a small handful of apps, including meme creator Whale, conversational app Bump, music app Aux, video app Hobbi, couples app Tuned, Apple Watch app Kit, and just yesterday, group calling app CatchUp. (Bump has since shut down.)

Prior to CatchUp, apps were launched with little fanfare but now Facebook is publically announcing their debuts and answering questions. That’s a change in strategy for the team, and one that could point Facebook’s desire to capitalize on users’ hunger for new social and entertainment experiences while stuck at home due to the COVID-19 outbreak.

Collab is available as an invite-only beta on iOS in the U.S. and Canada.

Spotify officially launches a shared-queue feature called Group Session

Spotify is today introducing a new feature, Group Session, that will allow two or more of its Premium users in the same space to share control over they music being played. Essentially a “party mode” of sorts, Group Session participants can control what’s playing in real-time as well as contribute to a collaborative playlist for the group.

The company notes that the feature can be used among quarantine-mates and families — groups that are now spending long hours at home together, where they now work, play, cook, dance, and more with Spotify running in the background.

To use the feature, the “host” will tap the Connect menu in the bottom-left corner of their Play screen, then share the scannable code that appears with their “guests.” The guests then join the session by scanning the host’s code. Then, using Spotify’s built-in controls, they’re able to pause, play, skip and select tracks on the queue and add in choices of their own to be played next. The changes any guest makes are immediately displayed to all participant’s devices.

Spotify already offers ways for groups to share their favorite music, but in more limited ways. The company offers a way to build Collaborative playlists with friends, where everyone can add, delete, and reorder tracks. In addition, Spotify Premium Family plan members can listen to a personalized playlist called Family Mix that combines the music that everyone enjoys.

However, neither of these options offer a way to collaborate in real-time, as Group Session does.

The Group Session feature has been in testing since last year, where it was first uncovered by noted reverse engineer, Jane Manchun Wong back in May 2019. Others had also reported seeing it appear on their own accounts later in August. That means some users may have had the feature before today. However, it’s only now being made globally available to all Premium users.

At the time of its initial development, Spotify may have envisioned the feature as a way to make its app go viral, as users would download Spotify in order to contribute to a party playlist — perhaps by scanning a code that appeared on the party host’s TV, for example. But with the coronavirus pandemic limiting gatherings and people isolating themselves at home, the company is instead positioning Group Session as a way for families and housemates to entertain themselves.


By tying a feature like this to its Premium subscription, Spotify is hoping to encourage more of its free users to make the jump to its paid, music-on-demand streaming service. It’s a good time for this sort of push, too.

More people are staying at home due to the coronavirus pandemic, and are looking for ways to be entertained. Streaming, particularly video, is up. And Spotify said during its April earnings it was seeing double-digit increases over the past few weeks in listening around in-home activities, like cooking, chores, family time, and more.

Today, Spotify says Group Session is launching into beta, meaning it’s still being tested and developed.

The version launching today is the first iteration of the feature, and Spotify expects it will evolve over time as it learns more from user feedback. It may even expand beyond people sharing the same space in the future.

Group Session is rolling out today, so you may not see it just yet. The feature will be available to all Premium users worldwide when the rollout completes.

OpenAI’s new experiments in music generation create an uncanny valley Elvis

AI-generated music is a fascinating new field, and deep-pocketed research outfit OpenAI has hit new heights in it, creating recreations of songs in the style of Elvis, 2Pac and others. The results are convincing, but fall squarely in the unnerving “uncanny valley” of audio, sounding rather like good, but drunk, karaoke heard through a haze of drugs.

Jukebox, the organization’s new music-generating system, was detailed in a blog post and paper published today. OpenAI produced some interesting work almost exactly a year ago with MuseNet, a machine learning system that, having ingested a great deal of MIDI-based music, was able to mix and match genres and instruments.

But MIDI is a simpler format than final recorded music with live instruments, since the former consists of discrete notes and key presses rather than complex harmonics and voices.

If you wanted an AI to examine the structure of a classical piano piece, the timing and key presses might only amount to a couple thousand pieces of information. Recorded audio is far denser, with (usually) 44,100 samples per second.

Machine learning systems that learn and imitate things like instruments and voice work by looking at the most recent words or sounds and predicting the next few, but they generally operate on the order of tens or a hundred pieces of data — the last 30 words or notes predict what the next 30 will be, for instance. So how can a computer learn how a tiny fraction of a waveform 10 seconds and 440,000 samples into a song compare with a sample 90 seconds and 4 million samples in?

OpenAI’s solution is to break down the song into more digestible parts — not quite key and chord, but something like that, a machine-palatable summary of 1/128th of a second of the song, picked from a “vocabulary” of 2,048 options. To be honest it’s hard to create an analogy because this is so unlike the way humans remember or understand things — as far as we even understand that.

It doesn’t actually use color swatches — that’s just to indicate that it’s breaking the waveform down into pieces.

The end result is that the AI agent has a reliable way to break down a song into digestible bits that are big enough that there aren’t too many to track, but small enough that they can reliably reconstruct the sound of a song. The process is much more complex than it sounds here; reliably breaking down a song to a series of “words” and then reconstructing it from them is the core of the new research, but the technical details I’ll let the OpenAI team to explain in their paper.

The system also had to learn how to parse the lyrics in a song, which like most things in this domain is more complicated than it sounds. Our ability to remember and use vocal patterns is partly innate and partly learned, and we tend to take for granted how powerful it is. Computers have no such ability and must learn how to pick out a voice from a mix, understand what it’s saying and match that to lyrics that are nothing more than a series of words with no information on key, tempo and all the rest. Nevertheless the OpenAI system does it to a satisfactory degree.

Jukebox is able to accomplish a variety of musical tasks, and while the results aren’t what you might call singing material, it must be kept in mind that there’s very little like this out there now, able to rebuild a song from scratch that’s recognizable as being like the target artist. Trained on 1.2 million songs, the system in the end has one multifaceted ability it accomplishes these tasks with: essentially, improvising a song given lyrics and the style it has learned from ingesting others by that artist.

So given its knowledge of how Ella Fitzgerald sings and the way instruments generally accompany her, it can sing a rendition of “At Long Last Love” in a way that sounds like her but definitely isn’t what Cole Porter had in mind. (Samples for these examples and more are included near the top of the OpenAI blog post.)

Jukebox can also sing entirely original lyrics in another’s style, like this truly strange Elvis song, “Mitosis,” written by another AI language model:

In case you didn’t catch that:

From dust we came with humble start;
From dirt to lipid to cell to heart.
With [mitosis] with [meiosis] with time,
At last we woke up with a mind.
From dust we came with friendly help;
From dirt to tube to chip to rack.
With S. G. D. with recurrence with compute,
At last we woke up with a soul.

Yes, it’s “Elvis” using cell division as a metaphor for life, as imagined by an AI. What a world we live in.

Lastly, there’s the “completion” task, where Jukebox learns (in addition to the base learning from its library) from the first 12 seconds of a song and uses that to generate the rest in a similar style. The switch from original to AI-generated sounds a bit like the ether just kicked in.

While MuseNet could be played with more or less in real time due to its lesser complexity, Jukebox is hugely computation intensive, taking hours to generate a single second of music. “We shared Jukebox with an initial set of 10 musicians from various genres… these musicians did not find it immediately applicable to their creative process,” the authors note dryly. Still, it’s fun and fascinating research and, given the current cadence, we can expect an even further improved version of the OpenAI music effort next April.

SiriusXM rises on Q1 earnings beat, but warns of coronavirus impacts to come

SiriusXM’s first-quarter 2020 earnings today painted a picture of what’s ahead for the music and entertainment service in light of the coronavirus outbreak. While the company surprised with both an earnings and revenue beat in the quarter ended March 2020, its satellite radio business also lost net subscribers due to declines in auto shipments, and the company spoke of further declines in ad sales and in customer responses to its marketing campaigns.

The company did manage to beat expectations in the quarter, reporting revenues of $1.95 billion, surpassing the Zacks Consensus Estimate by 2.63%. And it saw earnings per share of $0.07 (a profit of $293 million), beating the estimate of $0.05 per share, and up from the earnings of $0.03 per share a year ago.

But the earnings beat comes at a time when even SiriusXM isn’t sure of what the future holds for its business — it withdrew its full-year 2020 guidance, citing the still unknown potential impacts of the COVID-19 crisis.

Already there were hints of how that future may look, however. The declines in shipments from automakers offering paid trial subscriptions with a vehicle purchase led the company’s satellite radio business to lose 143,000 net subscribers in the first quarter. This is despite the addition of 69,000 self-paying subscribers, and saw SiriusXM ending the quarter with 34.8 million total subscribers.

The Pandora streaming music service, which SiriusXM owns, added 51,000 net new self-pay subscribers to its paid tiers, Pandora Plus and Pandora Premium. Pandora ended the quarter with over 6.2 million self-pay subscribers and 6.3 million total paying subscribers, including those who came in through other promotions.

Pandora ad revenue grew 4% year-over-year to reach $241 million in the quarter, which the company attributed to video programmatic and engagement-based video, its expansion of off-platform efforts and the fees from its AdsWizz platform — a 2018 acquisition.

But Pandora’s gross profit was down 5% year-over-year, to $105 million, as total costs of services grew, including those related to higher revenue share and royalties, customer service, billing expenses and more.

Then there were the expected declines related to the coronavirus’ early impact.

Though much of those troubles didn’t hit until March, SiriusXM warned that “auto sales, advertising, and customer responses to marketing campaigns all fell swiftly in the second half of March.”

That’s only a couple of weeks, mind you, which makes it seem like the company hasn’t really felt the full force of the pandemic on its subscriber growth, ad sales or total revenues.

“Automakers have idled plants, and dealers have closed their retail operations. New and used vehicle sales have declined sharply in recent weeks,” the company said in its earnings announcement.

Beyond that, the overall economy has taken a hit, with rising unemployment and other declines that could touch on SiriusXM’s business in other ways — including the cancellation of sporting events, postponing of concerts, travel declines and more.

“Unemployment is rising at historic rates as non-essential businesses have been closed and workers have been furloughed. Media spending by businesses has dropped sharply. To add to the uncertainty, it is unclear when an economic recovery could start and what a recovery will look like after this historic shutdown of the economy,” SiriusXM also warned.

The company says it expects to see declines in ad revenues at both SiriusXM and Pandora due to the COVID-19 pandemic, as well as declines in its satellite radio and accessory sales.

Despite all these issues, there are areas where SiriusXM could succeed, as the coronavirus quarantine stretches on — specifically, its exclusive entertainment offerings.

Today, there are already signs that people are looking for other options to keep themselves entertained in quarantine beyond just streaming Netflix endlessly. Nintendo is struggling to keep the Nintendo Switch in stock, thanks to hit games like Animal Crossing, for example, and even podcast listening is starting to recover from the initial coronavirus hit.

SiriusXM could easily cater to the growing demand for virtual events, where music and entertainment reaches consumers stuck at home.

The company has already done this to some extent — with its Ultra Music Festival, home performance from Garth Brooks, Home DJ series kicked off by Taylor Swift, broadcasts of nationwide events like the Jersey 4 Jersey benefit and more. The company also began streaming a COVID-19 news channel 24/7, in conjunction with NYU Langone Health. Plus, it still has Howard Stern…for now.

Plus, as quarantined consumers dig further into non-screen-based activities — like gardening, arts and crafts and cooking, among others — SiriusXM could establish itself as a service offering more than just your usual tunes and podcasts if it strikes the right tone with regard to its marketing efforts.

“Since the start of the global pandemic, our top priorities have been ensuring our employees’ safety and well-being, and continuing to support our subscribers and listeners by providing them the best entertainment, news, and information in the audio space. On both fronts, I’m pleased by our response,” said SiriusXM CEO Jim Meyer, in a statement. “We are streaming SiriusXM for free, and we have been in overdrive introducing new shows, channels and special virtual moments,” he said.

SiriusXM also invested $75 million in SoundCloud in the quarter, in the weeks before the pandemic hit, which allows the company to reach 140 million North American listeners across SiriusXM, Pandora and SoundCloud combined.

SiriusXM’s stock is up 2.58% following its Q1 results.

Pandora’s new Apple Watch app lets you leave your iPhone behind

Support for standalone streaming has come to Pandora’s Apple Watch app. The company today announced the official launch of its new standalone app for Apple Watch that lets you listen to music and podcasts on the go, even without your iPhone. The launch makes Pandora the first major third-party — meaning, first besides Apple Music — to offer a standalone app for Apple Watch.

To be clear, Pandora is not the first non-Apple music service to offer an Apple Watch app. Spotify notably debuted its own Watch app in 2018. Others, including SoundCloud, Napster, Deezer and more, also have Apple Watch experiences.

However, Spotify’s app still needs to be tethered to the iPhone in order to work. This has been a sore subject with a core group of Spotify’s customers — particularly those who want to enjoy music on their Apple Watch while exercising, for example, when carrying around an iPhone is more cumbersome. In some cases, these users have even defected to Apple Music, calling Spotify’s watch app a mere “remote control,” as it still hasn’t even implemented offline support. 

Pandora’s new app, on the other hand, lets users leave their phone behind as it supports both streaming and offline downloads.

The app notably takes advantage of the new streaming APIs Apple introduced last year at its annual developer conference. With watchOS 6, app developers can now create independent audio consumption experiences that no longer need to lean on the iPhone.

For Pandora, this change means users can go directly to the App Store on the Apple Watch to download the watch app to log in and start streaming. Even without an iPhone, users can play, pause and skip songs; pick up where they left off on podcasts; thumbs-up their favorite music; adjust the volume; and more.

Pandora Premium subscribers also can search and play specific songs, artists and albums on-demand, right from their wrist.

Meanwhile, offline listening is available to Pandora Plus or Premium paying subscribers, which lets you save tracks for offline play. This is helpful if you’re in an area where you have a poor connection or none at all, like on an airplane or underground train, for instance.

The new Pandora Apple Watch app will roll out to all users who already had the older version installed, as well as to those who updated to the latest version of Pandora’s iPhone app.

Because the app relies on Apple’s new streaming APIs to offer standalone streaming, it requires watchOS 6 to work. This version of Apple Watch software is available to Apple Watch Series 1 and higher users, but only those on Apple Watch Series 3 and higher will have access to standalone streaming, per Apple’s requirements. In addition, both streaming music and downloading require an internet connection, either Wi-Fi or cellular.

News of Pandora’s updated Watch app actually broke last week. But after a number of news reports announced its arrival, the company clarified it was only then rolled out to a small group (1%) of users at that time. Earlier this year, Pandora also rolled out a redesign of its Watch app that included offline playback.

As of today, the updated Apple Watch app should be available to all in the U.S.

Get ready to see Spotify’s looping videos on Instagram

Following Spotify’s confirmation of a new Stories feature, initially being tested by social media influencers, the company this morning announced it will now allow artists to reach their Instagram fan bases in a new way, too. However, in this case, they aren’t creating Spotify Stories they can market elsewhere on their social media, but instead are able to share their unique video art from Spotify’s Canvas feature directly to their Instagram.

Canvas launched into beta last fall, allowing artists to replace the album art that appears when a song is playing with a moving, visual experience that plays in a short loop. Canvas videos have had mixed reviews as some users find the imagery distracting while others seem to prefer it.

Starting today, the thousands of artists in the Canvas beta will be able to share their looping videos to Instagram with just a tap.

From the app’s Artists profile, each track that included a Canvas will have a “Share” icon next to it. By tapping that icon, artists can share the song and its Canvas to Instagram Stories. The post will look like a regular Spotify share with cover art and a link to play the track on Spotify. However, now their looping video will be the backdrop.

Currently, the Canvas beta is only available to those using the Spotify for Artists app on iOS. Spotify says it’s working to bring the sharing feature to Android users soon.

In addition, fans seeing the Canvas on Instagram aren’t counted in the Canvas metrics, unless they click through to Spotify, the company says.

The feature itself is intended to aid artists who are marketing their new songs to fans on Instagram as well as for highlighting updates to Canvas — like those that are updated to include clips from a new music video, new art, or live performances, for example.

One high-profile artist who’s taking advantage of Canvas is Billie Eilish — the artist who just swept last night’s Grammy Awards by winning the four biggest prizes — best new artist, record of the year, album of the year, and song of the year. Eilish has used Canvas to share animated versions of fan art, which helps her to better engage with her fan base.

Spotify claims that adding a high-quality Canvas has increased track shares by up to 200%, in addition to lifting streams, saves, and artist profile visits. By expanding Canvas to Instagram, those shares should bump up even higher, the company believes.

Despite its social media media-inspired features, like the new Stories addition or the looping videos of Canvas, Spotify doesn’t intend for its streaming app to become a new social platform. Instead, its focus is on building features that artists and listeners can leverage to better connect with social media fan bases elsewhere — either to help market themselves and their music or to improve discoverability of new music among their followers.

Artists interested in Canvas can sign up for the waiting list here.



Playing traffic cop for drones in cities and towns nets Airspace Link $4 million

As the number of drones proliferates in cities and towns across America, government agencies are scrambling to find ways to manage the oncoming traffic that’s expected to clog up their airspace.

Companies like Airmap and KittyHawk have raised tens of millions to develop technologies that can help cities manage congestion in the friendly skies, and now they have a new competitor in the Detroit-based startup, Airspace Link, which just raised $4 million from a swarm of investors to bring its services to the broader market.

The financing for Airspace Link follows the company’s reception of a stamp of approval from the Federal Aviation Administration for low-altitude authorization and notification capabilities, according to chief executive Michael Healander.

According to Healander, what distinguishes Airspace Link from the other competitors in the market is its integration with mapping tools used by municipal governments to provide information on ground-based risk.

“We’re creating the roads based on ground-based risk and we push that out into the drone community to let them know where it’s okay to fly,” says Healander.

That knowledge of terrestrial critical assets in cities and towns comes from deep integrations between Airspace Link and the mapping company ESRI, which has long provided federal, state and local governments with mapping capabilities and services.

We’ve just spent the past month understanding what regulation is going to be around to support it. In two years from now every drone will be live tracked in our platform,” says Healnder. “Today we’re just authorizing flight plans.”

As drone operators increase in number, the autonomous vehicles pose more potential risks to civilian populations in the wrong hands.

Parking lots, sporting events, concerts — really any public area — could be targets for potential attacks using drones.

“Drones are becoming more and more powerful and smarter,” EU Security Commissioner Julian King warned in a statement last summer, “which makes them more and more attractive for legitimate use, but also for hostile acts.”

Already roughly half of the population of the U.S. lives in controlled airspace where drones flying with more than a half a pound of weight require flight plan authorization, according to Healander.

“We build out population data and give state and local governments a tool to create advisories for emergency events or any areas where high densities of people will be,” says Healander. “That creates an advisory that goes through our platform to the drone industry.”

Airspace Link closed a $1 million pre-seed round in September 2019 with a $6 million post-money valuation. The current valuation of the company is undisclosed, but the company’s progress was enough to draw the attention of investors led by Indicator Ventures with participation from 2048 Ventures, Ludlow Ventures, Matchstick Ventures, Detroit Venture Partners and Invest Detroit.

For Healander, Airspace Link is only the latest entrepreneurial venture. He previously founded GeoMetri, an indoor GPS tracking company, which was acquired by Acuity Brands.

I’ve been a partner of ESRI my entire life,” says Healander. “I’ve been in the geospatial industry for four or five companies with them.”

The company has four main components of its service. There’s AirRegistry, where people can opt-in or out of receiving drone deliveries; AirInspect, which is a service that handles city and state permitting for drone operators; AirNetm, which works with the FAA to create approved air routes for drones; and AirLink, an API that connects drone operators with local governments and collects fees for registering drones.

Spotify’s new test lets influencers post Stories to introduce their own playlists

Spotify is testing a new Stories feature that will allow select influencers to incorporate video elements into their public playlists, TechCrunch has learned and Spotify confirmed. The first influencer to test the feature is makeup and fashion YouTube star Summer Mckeen, who currently has a social media fan base that includes 2.33 million YouTube subscribers, 2.1 million Instagram followers, and 126,455 Spotify followers. Mckeen is using the new feature to introduce a playlist of her all-time favorite songs, which she’s titled her “all time besties.”

Like other Stories’ products found on social media apps, the Spotify version offers a similar experience that includes short video clips that users can tap on to advance to the next screen. There are also horizontal lines at the top that indicate how many screens still await them ahead.

Above: where Stories are found on playlists

Meanwhile, the entry point for the Spotify Story is a circular icon right found above the playlist’s title. This has also been designed to catch your attention with an animated preview of the video you’ll see if you tap through.

Above: Mckeen introduces her playlist of favorite songs

Once in the Story, the clips will play and advance automatically and the playlist where you found the Story is featured at the top. You can also tap the “X” to exit at any time.

Spotify’s unique take on the Stories format involves its use of music, of course.

In the new Stories feature, the influencer can also share video clips that contain small song snippets and the album art as a way of previewing the songs in the playlist. In Mckeen’s case, a few of these follow her introduction of the new playlist.

Above: Song clips in Stories

Mckeen is the first influencer to go live on Spotify Stories, but we understand the company is also planning to roll this out to other notable names across the entertainment, lifestyle and music industries in the near future. This initial group of testers is being determined by a variety of factors — including follower count, how engaged their followers are, and how active the influencer in question is on Spotify. Mckeen was selected because she’s someone who likes to make playlists on her own and has many user-generated playlists she shares with fans.

Spotify isn’t rolling out the feature to its artists, however, as it’s meant to be more a tool for music discovery, rather than one for promotional purposes. Artists, instead, can reach fans creatively using Canvas — the recently launched looping videos product that can take the place of album art when a song plays.

Despite the similarity with other Stories found on apps like Instagram, Facebook, Snapchat and YouTube, Spotify’s goal isn’t to turn its app into another social media platform. Instead, it will rely on the influencers to get the word out to fans themselves using their existing accounts found elsewhere. Mckeen, for example, posted to her Instagram Stories with a deep link directly to the playlist in question.

Above: Mckeen’s IG Story

Currently, the Spotify Stories product can only be seen on iOS and Android, not desktop. (Mckeen’s is here.) And it’s available to all Spotify users — free and paid — although that could one day change. Spotify considers the product just a test for now, but is open to considering a broader rollout in the future.

Spotify confirmed the test to TechCrunch and offered a short statement.

“At Spotify, we routinely conduct a number of tests in an effort to improve our user experience. Some of those tests end up paving the path for our broader user experience and others serve only as an important learning,” a Spotify spokesperson said. “We have no further news to share on future plans at this time,” they added.

This is not the first time Spotify has dabbled with a Stories format, however. Last year, Spotify was spotted testing a Stories-like product called Storyline that was similar to “Behind the Lyrics,” but instead allowed artists to share their own insights, inspiration, and other details more directly. This can still be found on Spotify on select songs, but hasn’t become broadly available.

SiriusXM and Pandora test bundle discounts

It’s been less than a year since SiriusXM completed its $3.5 billion acquisition of streaming music service Pandora, but the two companies have already leveraged their collective assets to boost each other’s services. For example, SiriusXM talk shows arrived on Pandora as podcasts, while a Pandora-powered station now streams popular songs for both sets of listeners. Now, the company is considering tying the two services together in a different way — by packaging them as a discounted bundle.

What that bundle deal will look like isn’t yet known.

Pandora today offers four tiers of service: a free ad-supported version, the $4.99/month Pandora Plus service, and the $9.99/month Spotify rival Pandora Premium. It also offers a multi-user Pandora Premium Family plan for $14.99/month.

SiriusXM, meanwhile, also offers its own set of packages, with the most popular being a $5/month plan for the car and home (via an Echo device), an $8.25/month plan for in and out of the car, and an $8/month plan for streaming outside the car only.

Before rolling out a bundle deal, the company wanted to know what sorts of package price points and features customers would respond to best.

The company confirmed it’s been testing different cross-promotions, including those aimed at both Pandora and SiriusXM subscribers that offered discounts if you sign up for the other service. Essentially, the company wants to know what price point makes sense for consumers when it comes to subscribing to both services.

Today, these cross-promotions are aimed only at people who already subscribe to one or the other service, so it’s not really being marketed as a “bundle” deal yet. It’s just a promotion, if you want to get technical about the terminology.

“We would email our Pandora listener base or the SiriusXM listener base — we would test it with different user bases as a promotion,” Chris Phillips, SiriusXM/Pandora Chief Product Officer & Head of Technology, told TechCrunch. “We actually have a formal study going on to do it,” he said.

SiriusXM and Pandora haven’t yet settled on what a potential bundle deal will look like, but it aims to make a decision based on its tests this year.

“The power of the Sirius brand and power of the Pandora brand are very distinct. And people see unique value in the two,” Phillips added.

One challenge, however, is that people don’t understand that SiriusXM and Pandora are now one company, so the promotional emails confused them.

Similarly, people often find the language around “Pandora-powered” stations in SiriusXM confusing, as well.

One potential solution is to pick one consumer-facing brand and merge assets, including both programming and apps.

When asked if the two apps may merge into one in the future, Phillips said the company is “looking at what those opportunities might be.”

In the meantime, the company continues to explore how it can enhance both products using assets it has from the respective products.

“We are cross-pollinating content and features…into the distinct [user interfaces],” he said.

A recent example of this includes a new button within the SiriusXM app that allows you to launch a Pandora station based on what you’re currently streaming. And this new Pandora-powered station can then play right in the SiriusXM app — you don’t have to launch Pandora separately to hear it.

Efforts like this are aided by the fact that SiriusXM immediately put the two companies’ development groups together following the acquisition.

“We’re giving listeners choice. But when we give them choice, we want them to be able to have the best of what we offer in many places,” noted Phillips, of these sorts of integrations. “In the future, the idea that there’s a single opportunity — we’re looking at what that might be,” he said.

Nothing is yet determined, so all these plans could change, of course.

SiriusXM ended 2019 with around 30 million self-pay satellite radio and a record high of 34.9 million total paid subscribers. In 2020, SiriusXM forecasts revenue of $8.1 billion and earnings of $2.5 billion (adjusted EBITA).

Combined, Pandora and SiriusXM reach 100 million U.S. listeners per month.

YouTube Music adds three new personalized playlists, including its Spotify Discover Weekly rival

YouTube Music is taking on Spotify, Apple Music and others with the launch of three personalized playlists, including its own version of Spotify’s Discover Weekly, called Discover Mix, as well as a New Release Mix and Your Mix. Discover Mix had been spotted in the wild during testing, but now all three are globally available to YouTube Music users.

The company’s plans to introduce these new mixes were announced this fall at TechCrunch Disrupt SF 2019, where YouTube Chief Product Officer Neal Mohan spoke about the service’s plans to utilize a combination of machine learning and human curation to improve the music service’s offerings.

The Discover Mix is very much like Spotify’s Discover Weekly, as it will focus on helping users uncover new artists and music they like, including tracks you’ve never listened to before as well as lesser-known tracks from artists you already love. But unlike on competitor music services, this playlist can leverage historical listening data on both YouTube Music and on YouTube itself.

The mix, which updates every Wednesday, will give listeners 50 tracks per week.

The New Release Mix, as you can guess, focuses on all the recent releases by your favorite artists and others YouTube thinks you’ll like. This one drops every Friday, as most new releases do, but will add other tracks mid-week as needed.

Finally, Your Mix is a playlist that combines the music you love with songs you haven’t heard yet but will probably like, based on your listening habits. This one updates regularly to stay fresh.

Of course, the longer you listen on YouTube Music, the better the mixes will get. But YouTube says it can offer personalized mixes as soon as a user selects a couple of artists they like during the setup process or after they listen to a couple of songs.

The mixes arrive at a time when Google is more heavily investing in its streaming music service. Earlier this fall, it made YouTube Music the default music app that ships with new Android devices, instead of Google Play Music. And recently, reports indicate that YouTube Music is ahead of Spotify and JioSaavn in India, a key market for Spotify, despite its late entry.

The new mixes are live today on YouTube Music across iOS, Android, and the web.