UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
)
SENSORY, INC., )
)
Plaintiff, )
)
v. ) Case No. 24-cv-02788 (APM)
)
GOOGLE LLC, )
)
Defendant. )
_________________________________________ )
MEMORANDUM OPINION AND ORDER
I. INTRODUCTION
Before the court is another antitrust case brought against Defendant Google LLC that rests
in part on this court’s determination that Google is a monopolist in the general search services
market. Plaintiff Sensory, Inc. is a software company that develops and licenses software
associated with wake words and voice assistants. Wake words are verbal commands that activate
a voice-controlled device. Think “Hey Siri” or “Hey Google.” Sensory alleges that Google’s
exclusionary distribution agreements with Android device manufacturers violate Sections 1 and 2
of the Sherman Act and parallel provisions of the D.C. Code in nearly a dozen different product
markets. They include not only the markets for wake word and voice assistant technologies, but
also those for general search services, general search text advertising, access points to general
search services, and speech-to-text voice recognition software, among many others. Google moves
to dismiss, arguing that Sensory has failed either to plead antitrust standing or to adequately define
these product markets.
For the reasons that follow, Google’s Motion to Dismiss, ECF No. 20, is granted in part
and denied in part. Only those claims relating to the four alleged product markets for wake word,
voice assistant, and voice recognition technologies used to support Android smartphones and
tablets may proceed to discovery. Sensory’s Motion for Leave to File a Surreply in Opposition to
Defendant Google LLC’s Motion to Dismiss, ECF No. 23, is denied.
II. BACKGROUND
Sensory’s factual allegations include numerous references to the proceedings in
United States v. Google LLC (Google Search), 747 F. Supp. 3d 1 (D.D.C. 2024), including this
court’s liability determination, trial exhibits and testimony, and the parties’ filings. See, e.g.,
Pl.’s Renewed Mot. for Leave to File First Am. Compl. Partially Under Seal, ECF No. 19,
First Am. Compl., ECF No. 19-1 [hereinafter Am. Compl.], ¶¶ 9–10, 38–62, 70–71, 80–100, 176–
180, 189. The court therefore will consider the relevant factual findings made in that case.
Otherwise, the court will focus on those allegations unique to Sensory’s claims.
A. Wake Words and Voice Assistants
Wake words are words or phrases used to “initiate verbal communication” with a device.
Id. ¶ 15. On smartphones and tablets, a wake word is “commonly used to access voice-based
‘assistants,’ which are used to conduct internet searches and return information to the user.” Id.
¶ 181. A voice assistant “is a virtual assistant that can respond to voice commands to perform
various tasks.” United States v. Google LLC (Google Search Summ. J.), 687 F. Supp. 3d 48, 86
(D.D.C. 2023) (internal quotation marks omitted).
Each voice assistant is typically associated with a unique wake word. Am. Compl. ¶ 182.
For example, “Hey Siri” is the unique wake word activating Siri, the voice assistant on Apple
devices, and “Hey Google” or “OK Google” are the unique wake words for Google Assistant, the
2
voice assistant on certain Android devices. Id. ¶¶ 15, 63. But wake words also are used to initiate
verbal communication with devices other than smartphones and tablets. Id. ¶¶ 20, 30.
As discussed, voice assistants can be “used to conduct internet searches and return
information to the user.” Id. ¶ 181. When a user makes a spoken search query to the voice
assistant, the voice assistant will transcribe the spoken query and then enter the transcribed text
into a search engine. Id. ¶ 65. So “[f]rom the perspective of the search engine, the system runs
the same whether the search query is typed or spoken.” Id. Voice assistants can therefore be
considered search access points, which are “the places at which a consumer may enter a general
search query.” Id. ¶ 213; see also id. ¶¶ 63, 66, 197.
B. Sensory’s Products and Technologies
Sensory is a Silicon Valley–based software company that describes itself as an “innovator
in the development of wake words and other voice recognition technology, including custom voice
assistants, voice control, and sound ID technologies.” Id. ¶ 1. It claims to be the “first technology
company to solve certain specific problems in wake words and their deployment with voice
assistants,” including “(1) avoiding false acceptance (waking up when no one intended the device
to wake up); (2) avoiding false rejection when the consumer says the right word and the device
doesn’t respond; (3) addressing privacy concerns requiring the wake word technology to run
locally on device so spoken data is not constantly streamed to a cloud; and (4) keeping power
consumption to a minimum to reduce battery drain for devices that are not plugged in.” Id. ¶ 16.
It also claims to have “solved complicated technological problems related to a concept called
‘concurrency,’” which “allows more than one voice assistant to run simultaneously on a device
and to be activated by distinct wake words.” Id. ¶ 17.
3
Sensory develops and licenses three products related to wake words: TrulyHandsfree,
TrulyNatural, and TrulySecure. Id. ¶ 24. TrulyHandsfree “covers wake words and commands
using keyword spotting.” Id. Sensory alleges that, “[f]or certain customers, Sensory’s
TrulyHandsfree Product on its own acts [as] a voice assistant, because it can both activate a device
when it detects that the appropriate wake word has been spoken, and it can follow up device
activation with an action, such as ‘call,’ ‘take a photo[,]’ ‘take a video’ or a similar command.” Id.
¶ 25. TrulyNatural “supports larger vocabularies using grammars, speech to text, acoustic models
and language models,” id. ¶ 24, and can “enhance[] the voice assistant capabilities of a device,”
id. ¶ 26. And TrulySecure “permits speaker verification, Face ID, and sound ID.” Id. ¶ 24.
Sensory’s “wake word technology” has been licensed and used by several major original
equipment manufacturers (“OEMs”) of smartphones, tablets, and other hardware, including
Microsoft, Apple, Samsung, Motorola, Lenovo, and Google. Id. ¶¶ 19–20, 23, 27. Sensory alleges
that its wake word technology was once used on many smartphones and tablets running on the
Android operating system, id. ¶¶ 23, 28–29, and that its “wake words have been the front end of
many search engines[1] including Google, Cortana, Siri, and Alexa,” id. ¶ 21. Other leading
consumer electronics manufacturers, such as Amazon, Hasbro, Plantronics, GoPro, Tencent, and
Garmin, also have integrated Sensory’s technologies into their products. Id. ¶ 20.
C. Google’s Wake Words and Voice Assistants
In 2012, Google endeavored to incorporate wake words and spoken command capabilities
into a new product, Google Glass. Id. ¶ 30. But because Google did not at that time “have its own
1
The court presumes reference to “search engines” in paragraph 21 is an error, as Cortana, Siri, and Alexa are voice assistant products, not search engines. Google, of course, is a search engine, but given its inclusion alongside other voice assistants, Plaintiff likely here means Google Assistant. If Sensory’s wake word technology had at any time been the “front end” of the Google search engine, presumably Sensory would have said so clearly and precisely, and not as the outlier in a list of otherwise similar products.
4
wake word or voice assistant technology,” it licensed TrulyHandsfree from Sensory “for use of
specific wake words and [spoken] commands.” Id. Google was then “so impressed with Sensory’s
technology” that it entered into negotiations to acquire Sensory. Id. ¶ 32. An acquisition ultimately
did not occur, and Google thereafter began developing its own wake word and voice assistant
technologies for use on Android smartphones and tablets rather than license Sensory’s products.
See id. ¶¶ 32–37.
At the time Google first developed and used its own wake word 2 and voice assistant
technologies, device manufacturers could still “use Sensory’s wake word and voice assistant
technology to initiate search queries on a competing search engine.” Id. ¶ 68. But Google then
“began to distribute [its] technologies to virtually every smartphone and tablet sold in the United
States that is not made by Apple through modifications to its Mobile Device Distribution
Agreement (MADA). These modifications included: (1) requiring the use of Google [wake words]
and the Google Assistant and (2) in many cases to require that the use of Google [wake words] and
the Google Assistant be exclusive.” Id. ¶ 37; see also id. ¶ 69. Google’s revenue share agreements
(RSAs) also “included restrictions preventing alternative assistant[s] from being place[d] on the
home screen of a device and required that assistants provided from third parties (like Sensory) and
carriers could not be enabled ‘out of the box,’” imposing “cumbersome activation steps for the use
of non-Google assistants.” Id. ¶ 61.
From this, Sensory alleges, a stream of anticompetitive effects flowed:
First, because manufacturers were required to install Google Assistant as the exclusive,
default voice assistant on their devices, the MADAs and RSAs had the practical effect of
“shut[ting] out potential competitors to Google Assistant.” Id. ¶ 71. In other words, device
2
Google also refers to its wake words as “hotwords.” Am. Compl. ¶ 30.
5
manufacturers were prevented “from using non-Google wake words and voice assistants” and
“from using technology that would allow wake word/voice assistant concurrency.” Id. ¶¶ 72–73.
Second, because manufacturers could not use voice assistants other than Google Assistant
or wake words other than those associated with Google Assistant, device users’ spoken search
queries were also defaulted to Google Search. See id. ¶ 63 (alleging that Google Assistant is an
“access point[] to Google’s general search engine”). And even “[i]n instances where Google [did]
not own the consumer-facing voice assistant,” Google “contracted to ensure that the consumerfacing voice assistant relies upon Google’s general search services for general queries.
For instance, both Apple’s Siri and Samsung’s Bixby use Google’s general search services to
search the internet.” Id. ¶ 70. “Because voice assistants are an access point to a general search
engine, Google sought to control how and which voice assistants could be invoked . . . to protect
and extend its monopoly in general search engines by cutting off access to alternative services.”
Id. ¶ 66.
Third, by disallowing concurrent voice assistants on devices, Android users were precluded
from accessing “different voice assistants and different search engines outside of Google’s
control.” Id. ¶ 194. A device with concurrency capabilities would theoretically be able to operate
other voice assistants with their own unique wake words alongside Google Assistant on the same
device. Id. ¶ 202. But Google prevented concurrent voice assistants through both (1) the exclusive
distribution of Google Assistant, which accordingly “foreclosed or limited Sensory’s opportunity
to provide its wake word products, including its concurrent wake word product,” id. ¶¶ 202, 206,
and (2) contractually obligating OEMs to exclude non-Google wake words and voice assistants
from their devices, id. ¶¶ 204, 206, 209. As a result, devices were not able to “support and enable
6
competition in the search-related markets by mixing and matching different (and non-Google)
search engines for different purposes.” Id. ¶ 191.
“If free from Google’s anticompetitive conduct,” Sensory contends, “voice-based searches
present avenues for competition to Google’s dominance in search.” Id. ¶ 190. For Sensory,
Google’s conduct allegedly has resulted in a complete loss of revenue generated from “companies
that used Google’s Android operating system.” Id. ¶ 78. By 2022, TrulyHandsfree had
“disappeared completely” from smartphones and tablets. See id. ¶ 79.
D. Sensory’s Claims
Sensory advances three claims. First, Sensory alleges Google has violated Section 2 of the
Sherman Act by unlawfully maintaining monopolies in at least 11 different markets:
• “general search services,” id. ¶ 86;
• “general search text ads,” id.;
• “general search advertising,” id. ¶ 178;
• “access points to general search services on phones and tablets running the Android
operating system,” id. ¶ 226;
• “access points to general search services and voice searching on Android phones
and tablets,” id. ¶ 211;
• “voice recognition software (speech to text)” on Android smartphones and tablets,
id. ¶ 167;
• “voice recognition software (speech to text) used for search” on Android
smartphones and tablets, id. ¶ 228;
• “wake words on smartphones and tablets running the Android operating system,”
id. ¶ 152;
• “voice assistants on smartphones and tablets running the Android operating
system,” id. ¶ 154;
• “wake word technologies” on Android smartphones and tablets, id. ¶ 140; and
• “voice assistant technology on smartphones and tablets running the Google Android
operating system,” id. ¶ 141.3
3
Sensory refers at times to a market for “voice search,” which it appears to use interchangeably with the market for “voice search access points.” See, e.g., Am. Compl. ¶¶ 157, 220. To the extent that is what Sensory intended, the court discusses the access points markets in Section IV.B. But to the extent Sensory intended “voice search” to be a market distinct from that for “voice search access points,” the court’s analysis of the general-search-services-related markets in Section IV.A applies, since “[f]rom the perspective of the search engine, the system runs the same whether the search query is typed or spoken.” Id. ¶ 65.
7
Second, Sensory alleges Google is engaged in an unlawful tying agreement in violation of
Section 1 of the Sherman Act. Sensory contends that Google conditions “distribution of Android,
Google Cloud, its Google search services, including its mobile applications (e.g., Google Phonetop Search and the Android Market Client and Chrome Browser) and its . . . Google Play store” on
both (1) “us[ing] Google’s [wake words] and Google’s Google Assistant” and (2) “not includ[ing]
other [wake words] or voice assistants, voice recognition software, and voice search access
points.” Id. ¶ 220.
Finally, Sensory incorporates all the same factual allegations to allege violations of parallel
antitrust provisions in the D.C. Code. See id. ¶¶ 232–238 (citing D.C. Code §§ 28-4502, 28-4503,
28-4508).
III. LEGAL STANDARD
To survive a motion to dismiss, a complaint must “state a claim to relief that is plausible
on its face.” Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009) (citation omitted). A claim is plausible
on its face if it alleges “factual content that allows the court to draw the reasonable inference that
the defendant is liable for the misconduct alleged.” Id. The court must accept as true all factual
allegations contained in the complaint, Bell Atl. Corp. v. Twombly, 550 U.S. 544, 556 (2007), and
construe the complaint in the plaintiff’s favor, Hettinga v. United States, 677 F.3d 471, 476 (D.C.
Cir. 2012). Although a plaintiff need not provide detailed factual allegations, the plaintiff’s
“obligation to provide the grounds of his entitlement to relief requires more than labels and
conclusions.” Twombly, 550 U.S. at 555 (cleaned up). The court therefore must not accept “legal
conclusion[s] couched as . . . factual allegation[s],” Papasan v. Allain, 478 U.S. 265, 286 (1986),
or “inferences drawn by [the] plaintiff if those inferences are not supported by the facts set out in
the complaint,” Langeman v. Garland, 88 F.4th 289, 294 (D.C. Cir. 2023) (citation omitted).
8
In antitrust cases, the court must also bear in mind that, while “it is one thing to be cautious before
dismissing an antitrust complaint in advance of discovery,” it is “quite another to forget that
proceeding to antitrust discovery can be expensive.” Twombly, 550 U.S. at 558.
IV. DISCUSSION
Sensory alleges a staggering 11 markets in which Google has committed antitrust
violations. See supra Section II.D. Google argues that Sensory’s claims as to all 11 markets must
be dismissed for failure to either plead antitrust standing or define a relevant market. The court
agrees as to seven of the markets—general search services, general search text ads, general search
advertising, access points to general search services and to voice searching, wake words, and voice
assistants. But the court finds that, as to the remaining four technologies markets, Sensory has
plausibly alleged antitrust standing and a relevant market definition. The court first discusses the
markets in which Sensory’s Section 2 claims are dismissed and then addresses the surviving four
technologies markets. The court then turns to Sensory’s tying and D.C. Code claims.
A. General Search Markets
1. Antitrust Standing
Like all plaintiffs in federal court, a private antitrust plaintiff must demonstrate a “case or
controversy” under Article III. But it must also do more: a private antitrust plaintiff must also
plead sufficient facts to establish antitrust standing. See, e.g., Atl. Richfield Co. v. USA Petroleum
Co., 495 U.S. 328, 334 (1990).
A plaintiff has antitrust standing if it suffers an antitrust injury. Andrx Pharms., Inc. v.
Biovail Corp. Int’l, 256 F.3d 799, 812 (D.C. Cir. 2001) (citing 2 Phillip E. Areeda et al., Antitrust
Law ¶ 337a (2d ed. 2000)). Antitrust injury is “injury of the type the antitrust laws were intended
to prevent and that flows from that which makes defendants’ acts unlawful.” Brunswick Corp. v.
9
Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977); see also Zenith Radio Corp. v. Hazeltine
Rsch., Inc., 395 U.S. 100, 125 (1969) (a cognizable antitrust injury is “the type of loss that the
claimed violations . . . would be likely to cause”); 2A Phillip E. Areeda & Herbert Hovenkamp,
Antitrust Law ¶ 337a (5th ed. & Suppl. 2026) [hereinafter Areeda & Hovenkamp] (“[A]n antitrust
violation can cause many types of injury, but only some of these are antitrust injury.”). It is not
enough that the injury is simply “causally related to an antitrust violation,” Atl. Richfield, 495 U.S.
at 334; the injury “should reflect the anticompetitive effect either of the violation or of
anticompetitive acts made possible by the violation,” Brunswick, 429 U.S. at 489. In other words,
“the antitrust standing inquiry turns on whether the plaintiff is a participant in the relevant market
and ‘suffered its injury in the market where competition is being restrained.’” Fotobom Media,
Inc. v. Google LLC, 719 F. Supp. 3d 33, 44 (D.D.C. 2024) (quoting Am. Ad Mgmt., Inc. v. Gen. Tel.
Co. of Cal., 190 F.3d 1051, 1057 (9th Cir. 1999)); accord In re Aluminum Warehousing Antitr.
Litig., 833 F.3d 151, 158 (2d Cir. 2016). Conversely, “[p]arties whose injuries, though flowing
from that which makes the defendant’s conduct unlawful, are experienced in another market do
not suffer antitrust injury.” FTC v. Qualcomm Inc., 969 F.3d 974, 992 (9th Cir. 2020) (quoting
Am. Ad Mgmt., 190 F.3d at 1057). Thus, “consumers and competitors are most likely to suffer
antitrust injury.” Am. Ad Mgmt., 190 F.3d at 1057.
Other market participants also can suffer antitrust injury. See id.; Mandeville Island Farms
v. Am. Crystal Sugar Co., 334 U.S. 219, 236 (1948) (The Sherman Act “does not confine its
protection to consumers, or to purchasers, or to competitors, or to sellers.”). But such participants
must still be comparable to “a customer who obtains services in the threatened market or a
competitor who seeks to serve that market,” SAS of P.R., Inc. v. P.R. Tel. Co., 48 F.3d 39, 44 (1st
Cir. 1995), for an injury that is “too secondary and indirect” from the anticompetitive conduct does
10
not suffice to establish antitrust standing, Serfecz v. Jewel Food Stores, 67 F.3d 591, 597 (7th Cir.
1995).
Injury that is “inextricably intertwined” with the injury suffered by competitors also may
be recognized “where the plaintiff was ‘used as a conduit to harm the defendants’ actual
competitors,’ such that the plaintiff’s harm is ‘an indispensable aspect of the scheme.’” Fotobom,
719 F. Supp. 3d at 47 (cleaned up) (quoting Hanover 3201 Realty, LLC v. Vill. Supermarkets, Inc.,
806 F.3d 162, 173 (3d Cir. 2015)). But even then, “harm that is secondary to the anticompetitive
conduct cannot support antitrust injury.” Hanover 3201 Realty, 806 F.3d at 173 (discussing
Blue Shield of Va. v. McCready, 457 U.S. 465 (1982)).
2. Sensory’s Antitrust Standing as to General Search Markets
The court first discusses the three alleged markets specific to general search that were at
issue in Google Search: general search services, general search text ads, and general search
advertising (collectively, “general search markets”).
Sensory alleges no facts establishing it suffers antitrust injury in the general search markets.
Sensory acknowledges that it “does not compete in the markets for general search services or
search-based advertising.” Pl.’s Opp’n to Def.’s Mot. to Dismiss, ECF No. 21 [hereinafter Pl.’s
Opp’n], at 1. In fact, it concedes that it does not participate in those markets in any capacity.
See Def.’s Mot. to Dismiss, ECF No. 20 [hereinafter Def.’s Mot.], at 9–10. See generally
Pl.’s Opp’n. Still, Sensory argues, “[t]he point is that those broader markets suffered a reduction
in competition based on the same conduct that directly harmed Sensory: by restricting the use of
competing wake words and voice assistants through its anticompetitive agreements, Google also
reduced competition in the broader general search services or search-based advertising markets by
controlling voice-based access points.” Pl.’s Opp’n at 1; see also id. at 2 (describing the “gravamen
11
of the Complaint” as being that “Google used the same exclusionary contracts the Court found
illegal not only to eliminate competition in the markets for wake words and voice assistants, but
also to reduce competition in general search services by controlling voice-based search access
points”).
But even if the court accepts as true that Google’s anticompetitive conduct as to wake
words and voice assistants also created anticompetitive effects in the general search markets, that
does not confer on Sensory antitrust standing in those markets. Sensory relies on a distorted
reading of Brunswick to argue that “Sensory can also base its action on other anticompetitive
effects made possible by Google’s restrictive agreements.” See Pl.’s Opp’n at 23. Brunswick
established the foundational principle that a private antitrust plaintiff’s “injury should reflect the
anticompetitive effect either of the violation or of anticompetitive acts made possible by the
violation.” 429 U.S. at 489. But that language was meant to narrow the world of cognizable
antitrust injuries, not expand it in the way Sensory contends. Cf. 2A Areeda & Hovenkamp ¶ 337a
(“This test forces antitrust courts to connect the alleged injury to the purposes of the antitrust laws.
. . . [A]n antitrust violation can cause many types of injury, but only some of these are antitrust
injury.”). It does not support Sensory’s proposition that a plaintiff who suffers injury in one market
can claim antitrust injury in a different market simply because the purported injuries can be traced
to the same anticompetitive conduct. The plaintiff must still be a participant in that market in
which it alleges an injury. See Fotobom, 719 F. Supp. 3d at 44 (citing Am. Ad Mgmt., 190 F.3d at
1057). Sensory has not pleaded that it is.
Sensory cannot overcome this deficiency by alleging that “causing harm and injury to
Sensory by pushing it out of the markets is an indispensable and intertwined part of Google’s
scheme to maintain dominance in, and control over, each of these markets.” Am. Compl. ¶ 162;
12
see also Pl.’s Opp’n at 25–26. Sensory appears to rely on Blue Shield of Virginia v. McCready,
which held that private plaintiffs that are not participants in the relevant market may nevertheless
suffer antitrust injury when the injury is “inextricably intertwined” with the injury the defendants
sought to inflict on market participants. See 457 U.S. at 483–84. But that holding applies narrowly
to cases where “the plaintiff was used as a conduit to harm the defendants’ actual competitors, such
that the plaintiff’s harm is an indispensable aspect of the scheme.” See Fotobom, 719 F. Supp. 3d
at 47 (cleaned up) (collecting circuit-court cases); accord 2A Areeda & Hovenkamp ¶ 339f
(“Although [‘inextricably intertwined’] is very elastic, the meaning was clearly limited in
McCready to those whose injuries are the essential means by which defendants’ illegal conduct
brings about its ultimate injury to the marketplace.”).
Sensory comes “nowhere close to clearing this high bar.” See Fotobom, 719 F. Supp. 3d
at 47. It attempts to bridge the gap between its injury and the injuries suffered by Google’s
competitors in the general search markets by seizing on the pleading shortcomings in Fotobom
Media, Inc. v. Google LLC. In Fotobom, this court held that the plaintiff, a manufacturer of smart
keyboards, had not plausibly alleged that its injury was “inextricably intertwined” with
anticompetitive effects in the general search services market because it “ha[d] not, for instance,
alleged that a meaningful percentage of search queries come through smart keyboards, such that
redirecting search traffic to, say, Bing would increase competition in general search.” Id.
(emphasis added). Sensory points out, parroting this language from Fotobom, that its complaint
does allege that “a meaningful percentage of search queries are conducted through voice search.”
Pl.’s Opp’n at 26 (citing Am. Compl. ¶¶ 186–189); see also Am. Compl. ¶ 161 (“[A] meaningful
and increasing percentage of search queries come through voice and are originated by wake words,
voice recognition software, voice searching, and related access points such that redirecting voice
13
search traffic to, say, Bing would have increased competition in general search and voice search.”).
But, even accepting that allegation as true, that is still a far cry from rendering plausible that
Google used Sensory as a “conduit” to harm its competitors in the general search markets.
Sensory’s conclusory allegations that its products “would have increased competition” in the
general search markets, see Am. Compl. ¶ 161, or that, “if consumers were to begin relying on
voice search not controlled by Google, those devices could support and enable competition in the
search-related markets,” id. ¶ 191 (emphasis added), are unavailing. See also Fotobom, 719
F. Supp. 3d at 47. Sensory pleads no facts showing that its injury “was necessary to [Google’s]
plan,” see Hanover 3201 Realty, 806 F.3d at 174, or was the result of the “manipulation of the
injured party as a means to carry out the restraint of trade in the product market,” Province v.
Cleveland Press Publ’g Co., 787 F.2d 1047, 1052 (6th Cir. 1986), such that it was “intertwined”
with the anticompetitive effects in the general search markets. At most, Sensory’s alleged injury
is no more than “incidental” or a “byproduct” of Google’s anticompetitive conduct. See In re
Aluminum, 833 F.3d at 161 (quoting Hanover 3201 Realty, 806 F.3d at 173–74). It is “secondary
to the anticompetitive conduct” in the general search markets and therefore “cannot support
antitrust injury.” See Hanover 3201 Realty, 806 F.3d at 173; see also SAS of P.R., 48 F.3d at 46
(reading McCready to apply even more narrowly to a plaintiff who is a “purchaser in the very
market directly distorted by the antitrust violation”).
Sensory’s other comparisons to Fotobom likewise fall flat. Sensory insists that, “[u]nlike
Fotobom,” its complaint “specifically alleges that Sensory’s injury also led to a reduction in
competition in the general search services market and the voice search market” and “alleges that
increased use of Sensory’s technology would enable other search engines to better compete with
Google or would facilitate new market entrants by reducing barriers to entry.” Pl.’s Opp’n at 25.
14
But these allegations are conclusory; and even if they were supported with well-pleaded facts,
establishing some loose connection between anticompetitive conduct in one market and harm in
another is not alone sufficient to confer antitrust standing.
Sensory is precisely the kind of party “whose injuries, though flowing from that which
makes the defendant’s conduct unlawful, are experienced in another market.” Qualcomm,
969 F.3d at 992. Such a party “do[es] not suffer antitrust injury” and therefore does not have
antitrust standing. See id. Sensory’s claims are dismissed insofar as they relate to the general
search services, general search text advertising, and general search advertising markets.
B. Access Points Markets
Next, the access points markets. Sensory alleges harms in the markets for “access points
to general search services on phones and tablets running the Android operating system,” id. ¶ 226,
and “access points to general search services and voice searching on Android phones and tablets,”
id. ¶ 211. Sensory’s claims as to these markets fail because it has not plausibly defined either.
“A threshold step in any antitrust case is to accurately define the relevant market.”
Coronavirus Rep. v. Apple, Inc., No. 21-cv-5567, 2021 WL 5936910, at *6 (N.D. Cal. Nov. 30,
2021). The relevant market is “the area of effective competition.” Qualcomm, 969 F.3d at 992
(quoting Ohio v. Am. Express Co., 585 U.S. 529, 543 (2018)); see also Image Tech. Servs., Inc. v.
Eastman Kodak Co., 125 F.3d 1195, 1202 (9th Cir. 1997) (“The relevant market is the field in
which meaningful competition is said to exist.” (citing United States v. Continental Can Co., 378
U.S. 441, 449 (1964))). “[W]ithout a definition of the market[,] there is no way to measure the
defendant’s ability to lessen or destroy competition.” Am. Express, 585 U.S. at 543. So “[t]o
survive a Rule 12(b)(6) motion to dismiss, an alleged product market must bear a ‘rational relation
to the methodology courts prescribe to define a market for antitrust purposes—analysis of the
15
interchangeability of use or the cross-elasticity of demand,’ and it must be ‘plausible.’” Todd v.
Exxon Corp., 275 F.3d 191, 200 (2d Cir. 2001) (citation omitted).
Nowhere does Sensory define the search access point markets with reference to
interchangeability of use or the cross-elasticity of demand. See id. That alone is reason to dismiss
Sensory’s claims as to these markets. See Chapman v. N.Y. State Div. for Youth, 546 F.3d 230,
237–39 (2d Cir. 2008).
But more fundamentally, search access points are not a product. Search access points are
a functionality that facilitates access to a search engine, such as through a browser feature or
application. Am. Compl. ¶¶ 45, 63; see also Google Search, 747 F. Supp. 3d at 44 ¶ 58 (describing
“search access points” as “channels of distribution” for general search engines to users, such as
browser search bars, widgets, applications, bookmarks, and direct web searches). And a market
inclusive of all search access points would be too broad to be plausible, as a desktop browser
search bar, for example, is clearly not reasonably interchangeable with a mobile widget. Cf. id. at
44 ¶ 58, 45 ¶ 64, 60 ¶ 155 (treating differently mobile and desktop search access points). Sensory’s
claims as to these markets must therefore also be dismissed.
C. Wake Words, Voice Assistants, and Technologies Markets
The court now considers the six markets related to wake words, voice assistants, and their
associated technologies on Android smartphones and tablets:
• “wake words on smartphones and tablets running the Android operating system,”
Am. Compl. ¶ 152;
• “voice assistants on smartphones and tablets running the Android operating
system,” id. ¶ 154;
• “wake word technologies” on Android smartphones and tablets, id. ¶ 140; and
• “voice assistant technology on smartphones and tablets running the Google Android
operating system,” id. ¶ 141
• “voice recognition software (speech to text)” on Android smartphones and tablets,
id. ¶ 167; and
16
• “voice recognition software (speech to text) used for search” on Android
smartphones and tablets, id. ¶ 228.
The court must first attempt to get its arms around what Sensory’s products are.
Sensory’s complaint is confusing. As discussed, Sensory alleges that it offers three
products—TrulyHandsfree, TrulyNatural, and TrulySecure. But Sensory waffles in how it
describes them. At times, TrulyHandsfree is a “software technology” that can facilitate a device’s
“listen[ing] in the background for a wake[] word without false firing and without any misses yet
not consuming too much power.” Am. Compl. ¶¶ 12–13; see also, e.g., id. ¶¶ 14, 16 (describing
TrulyHandsfree as a software that provides solutions to problems associated with wake words and
voice assistants); id. ¶¶ 19–20 (referring to its products as “technology” or “technologies”); id.
¶¶ 33–34 (referring to “wake word technology”). And it alleges that Google’s having developed
“its own wake word and voice assistant technology” pushed Sensory out of the market for those
technologies. Id. ¶ 37. But at others, one or more of Sensory’s products are themselves wake
words or voice assistants. See, e.g., id. ¶ 21 (referring to “Sensory’s wake words”); id. ¶ 25
(“For certain customers, Sensory’s TrulyHandsfree Product on its own acts a voice assistant.”); id.
¶ 33 (alleging that Google “began competing with Sensory in the market for wake words” when
Google began “develop[ing] its own hotword/wake word technology”). What’s more, it is not
always clear when or whether Sensory is referring to all three of its products or only
TrulyHandsfree, which Sensory invokes most often and explicitly describes as “cover[ing] wake
words and commands.” Id. ¶ 24. Sensory’s opposition offers little additional clarity. See, e.g.,
Pl.’s Opp’n at 15–18 (referring repeatedly to “wake word technology” and “the technology needed
to make” wake words as the product that Google originally licensed from Sensory and later
developed on its own, but simultaneously referring to the “market for wake words” as the affected
market).
17
Nevertheless, Sensory contends that it competes with Google in all six of these markets.
See id. at 18 (“[A]fter Google copied Sensory’s technology, it competed with Sensory in the market
to supply the technology needed to make [wake words] work, and then used its anticompetitive
MADAs and RSAs to eliminate Sensory as an alternative supplier of [wake words].” (citing Am.
Compl. ¶¶ 36, 37, 74–79)); id. at 22–23 (“Google . . . developed its own []wake word and voice
assistant technologies and began competing with Sensory in those markets.” (citing Am. Compl.
¶¶ 33–36)); id. (“Google unlawfully excluded its competitor, Sensory, from the markets for wake
words and voice assistants.”); Am. Compl. ¶ 228 (“Sensory was a participant (and a competitor of
Google) in the market for voice recognition software (speech to text) used for search on
smartphones and tablets running the Android operating system supplied by Google.”). And in a
section expressly titled “Additional Specific Allegations Relating to Antitrust Standing,” Sensory
alleges that Google “engage[s] in anticompetitive conduct that directly injured Sensory” in these
six markets. See id. ¶¶ 139, 141, 148, 149, 155, 156. Sensory does not allege it participates in
these markets in any other capacity.
As to the markets for wake words and voice assistants themselves, though skeptical, the
court will take Sensory at its word that these markets exist and even assume without deciding that
Google competes in them. See Am. Compl. ¶¶ 139–142. But even if Google is “engage[d] in
anticompetitive conduct” in those markets, Sensory has not in turn adequately pleaded that it
“suffered its injury in the market[s] where competition is being restrained.” Am. Ad Mgmt., 190
F.3d at 1057. As to the four technologies markets, the court finds that Sensory has plausibly alleged
antitrust standing and defined those markets.
18
1. Voice Assistant Market
Start with the market for voice assistants. Even taking the most generous interpretation of
any one of Sensory’s products or combination of products, none competes with Google Assistant,
Google’s voice assistant on Android devices.
Google Assistant “is a virtual assistant that can respond to voice commands to perform
various tasks,” Google Search Summ. J., 687 F. Supp. 3d at 86 (internal quotations omitted),
including “conduct internet searches and return information to the user,” Am. Compl. ¶ 181.
Google Assistant is therefore a search access point, where “a wake word that results in the
invocation of ‘Google Assistant’” allows a user to “use Google to conduct a search in particular.”
Id. ¶ 82. Today, Google Assistant also “incorporates LLM technology and GenAI functionality.”
United States v. Google LLC (Google Search Remedies), 803 F. Supp. 3d 18, 50 ¶ 18 (D.D.C.
2025).
At bottom, Sensory does not plausibly allege that it offers a voice assistant that resembles
Google Assistant. For “certain customers” that are never identified, Sensory alleges that its
“TrulyHandsfree Product on its own acts [as] a voice assistant, because it can both activate a device
when it detects that the appropriate wake word has been spoken, and it can follow up device
activation with an action, such as ‘call,’ ‘take a photo[,]’ ‘take a video[,]’ or a similar command.”
Am. Compl. ¶ 25. But that is all. Nowhere does Sensory allege that its voice assistant does
anything more—most notably, it does not allege that its voice assistant enables a user to conduct
internet searches on Android devices.
This is important. “[T]he relevant market must include all products ‘reasonably
interchangeable by consumers for the same purposes.’” United States v. Microsoft Corp., 253 F.3d
34, 52 (D.C. Cir. 2001) (emphasis added) (quoting United States v. E. I. du Pont de Nemours &
19
Co., 351 U.S. 377, 395 (1956)); see also Brown Shoe Co. v. United States, 370 U.S. 294, 325
(1962) (“The outer boundaries of a product market are determined by the reasonable
interchangeability of use or the cross-elasticity of demand between the product itself and
substitutes for it.”). So, “[i]n order to hold that [Sensory] was in competition with [Google],” the
court must be able to “conclude that what [Sensory] offered was reasonably interchangeable with
what [Google] offered.” See Barton & Pittinos, Inc. v. SmithKline Beecham Corp., 118 F.3d 178,
182 (3d Cir. 1997) (Alito, J.).
The court cannot so conclude, because Sensory has not alleged that consumers would use
any of Sensory’s products and Google Assistant “for the same purposes.” Namely, users cannot
use Sensory’s products as an access point for general search. Perhaps a user could use either
TrulyHandsfree or Google Assistant to ask a device to “‘call,’ ‘take a photo[,]’ ‘take a video[,]’ or
a similar command.” See Pl.’s Opp’n at 21–22. And perhaps, if considering all of Sensory’s
products together, Google Assistant also offers the functions that TrulyNatural and TrulySecure
purport to. Id. at 22. But even if two products may be reasonably interchangeable “for a discrete
purpose,” that does not mean they are reasonably interchangeable “for the same purposes.”
See Google Search, 747 F. Supp. 3d at 114.
FTC v. Whole Foods Market, Inc. and FTC v. Staples, Inc. illustrate this point further.
In Whole Foods, the D.C. Circuit recognized that “premium, natural, and organic supermarkets
(‘PNOS’)” like Whole Foods could make up a distinct market. 548 F.3d 1028, 1032, 1040
(D.C. Cir. 2008). It held so notwithstanding the fact that customers may “cross-shop” between a
PNOS and a conventional supermarket and find the same items there, because consumers
ultimately did not find PNOS “reasonably interchangeable” with conventional supermarkets.
See id. at 1040. And likewise in Staples, the district court held that office supply superstores
20
constituted a relevant product market even though consumers purchased office supplies at other
kinds of retail outlets, because there was a “low cross-elasticity of demand between the
consumable office supplies sold by the superstores and those sold by other sellers.” 970 F. Supp.
1066, 1078 (D.D.C. 1997). Thus, “the mere fact that a firm may be termed a competitor in the
overall marketplace does not necessarily require that it be included in the relevant product market
for antitrust purposes.” Id. at 1075; see also Whole Foods, 548 F.3d at 1040 (“But the fact that
[two firms] ‘are direct competitors in some submarkets . . . is not the end of the inquiry . . . .’”
(quoting United States v. Conn. Nat’l Bank, 418 U.S. 656, 663 n.3 (1974))). So even if Sensory’s
voice assistant and Google Assistant have some overlapping uses, “without more,” Sensory has
not shown “they belong in the same product market.” See Google Search, 747 F. Supp. 3d at 115.
Sensory does not dispute that its products do not offer search functionality. But it urges
that a voice assistant need not perform search functions. See Pl.’s Opp’n at 21–22. Fair enough,
technologically. But for antitrust purposes, a market that includes any “virtual assistant that can
respond to voice commands to perform various tasks” on an Android device is overbroad.
See Google Search Summ. J., 687 F. Supp. 3d at 86; see also Times-Picayune Publ’g Co. v. United
States, 345 U.S. 594, 612 n.31 (1953) (“A relevant market cannot meaningfully encompass [an]
infinite range [of products].”). Sensory offers no reason to think that there is any meaningful crosselasticity of demand between a voice assistant that can only perform tasks like “call,” “take a
photo,” or “take a video” and Google Assistant, which can do far more, or that consumers would
find such voice assistants reasonably interchangeable. See Brown Shoe, 370 U.S. at 325 (“[W]ithin
[a] broad market, well-defined submarkets may exist which, in themselves, constitute product
markets for antitrust purposes.” (citing United States v. E. I. du Pont de Nemours & Co., 353 U.S.
586, 593–95 (1957))).
21
Finally, Sensory’s singular allegation that “Sensory’s wake words have been the front end
of many [voice assistants] including Google, Cortana, Baidu, Siri, and Alexa” does not carry the
day. See Pl.’s Opp’n at 22 (citing Am. Compl. ¶ 21).4 That allegation only “states that Sensory’s
wake word technology has been used by other voice assistants, such as ‘Cortana, Siri, and Alexa,’
each of which themselves possess (separate) search functionality.” Def.’s Reply Mem. of P. & A.
in Supp. of Def.’s Mot., ECF No. 22 [hereinafter Def.’s Reply], at 9. It does not amount to an
allegation that TrulyHandsfree itself ever had search functionality.
Because Sensory has not alleged that its products are “reasonably interchangeable” with
Google Assistant, even if Google through its anticompetitive MADAs and RSAs caused harm in
the voice assistant market, that is not a market in which Sensory competes. See Am. Ad Mgmt.,
190 F.3d at 1057. Sensory does not suffer antitrust injury in the voice assistant market and
therefore does not have antitrust standing in that market.
2. Wake Word Market
Sensory’s argument that it suffers antitrust injury in the market for “wake words on
smartphones and tablets running the Android operating system supplied by Google” is even more
dubious. Besides intermittently referring to “Sensory’s wake words,” see, e.g., Am. Compl. ¶ 21,
or the “market for wake words,” id. ¶ 33, in which it alleges it competes, nowhere does Sensory
ever identify what its wake word is or was. And it does not identify any voice assistant on Android
devices that could be invoked by any of its alleged wake words.
In any event, no wake word that Sensory could offer could compete with Google’s wake
words. Google Assistant’s wake words are unique and specific to Google Assistant—“Hey
Google” or “OK Google”—just as Siri’s wake word—“Hey Siri”—is unique to it. Id. ¶¶ 15, 63.
4
The court notes that paragraph 21 of the complaint does not identify “Baidu” among the voice assistants for which Sensory’s technology has been the “front end.”
22
It is unclear how these may be “reasonably interchangeable” or how there could be a “crosselasticity of demand” for such words.5 Therefore, to the extent there is a market for wake words
in which Google competes, Sensory has not shown that it suffers antitrust injury in it. See Am. Ad
Mgmt., 190 F.3d at 1057.
3. Technologies Markets
Sensory’s claims as to the technology markets related to wake words and voice assistants,
including the voice recognition software markets (for search or otherwise), are different. Drawing
all reasonable inferences in Sensory’s favor, the court here assumes, as Google does, see Def.’s
Mot. at 14–15; Def.’s Reply at 3–4, that Sensory’s references to its “voice assistant” or “wake
words” are actually references to the technology underlying voice assistants and wake words. The
court also reads Sensory’s products, individually or collectively, as such technology.
Sensory alleges that, before the MADAs and RSAs bound OEMs to exclusive use of
Google Assistant and Google wake words, Google “faced competition in the wake word and voice
assistant [technologies] markets” for “smartphones and tablets running Google’s Android
operating system.” Am. Compl. ¶ 68. At that time, OEMs “could use Sensory’s wake word and
voice assistant technology to initiate search queries on a competing search engine.” Id. Indeed,
Sensory’s technology was used in “the front end of many [voice assistants] including Google
[Assistant], Cortana, Siri, and Alexa.” Id. ¶ 21. It was only after Google’s exclusive distribution
agreements that Sensory’s technology began disappearing from Android smartphones and tablets.
Id. ¶¶ 74–79.
5
Though the court assumes the existence of a market for wake words, Sensory also does not explain how the word required to communicate with a voice assistant is a “product” that can be “sold or licensed” independent from the voice assistant it invokes. See Def.’s Mot. at 16 (quoting Coronavirus Rep. v. Apple Inc., No. 21-cv-5567, 2021 WL 5936910, at *12 (N.D. Cal. Nov. 30, 2021)).
23
Accepting these allegations as true, it is plausible that (1) at the time Sensory was licensing
its products to OEMs of Android smartphones and tablets, Sensory’s products were reasonably
interchangeable with Google’s and (2) Google’s exclusive distribution agreements caused
anticompetitive effects in the market for those technologies. That is enough to survive a motion
to dismiss.
Google argues primarily that these claims fail because the technologies markets as alleged
are “implausible” simultaneously for being both over- and under-inclusive. Def.’s Mot. at 18–20.
First, Google characterizes Sensory’s alleged market definition as one consisting of “[a]ll
‘technologies’ that supposedly support assistant, wake word, and voice recognition functionality.”
Id. at 18. A market that includes all of TrulyHandsfree, TrulyNatural, and TrulySecure, Google
contends, is not one comprised of interchangeable products because these products have “vastly
different functions.” Id. And a market definition that consists of such “non-interchangeable
products” is impermissibly “vague and overbroad.” Id. at 18–19.
But it is Google’s reading of Sensory’s alleged technologies markets that is overbroad.
Sensory does not allege a market consisting of “all technologies” supporting voice assistants, wake
words, and voice recognition functionality. Rather, it specifies that the alleged market consists of
such technologies only to the extent they support those functionalities on Android smartphones
and tablets. See Am. Compl. ¶¶ 140, 141, 167, 228. And the allegations that its products could be
used to initiate search queries and were used as the front end of voice assistants—the “same
purposes” that Google’s wake word and voice assistant technologies have, see supra Section
IV.C.1—bear enough of a “rational relation to the methodology courts prescribe to define a market
for antitrust purposes.” Todd, 275 F.3d at 200.
24
But Google simultaneously thinks these markets too narrow. As to this argument, Google
acknowledges that “Sensory attempts to define markets for wake word, assistant, and voice
recognition technologies that are integrated into Android phones and tablets.” Def.’s Mot. at 19
(emphasis added). But, it continues, “Sensory pleads no basis to carve its market definition so as
to exclude the same technologies integrated into non-Android phones and tablets.” Id. Because
Sensory markets its technologies for use on non-Android devices and devices other than
smartphones and tablets, “any market definition which excludes such products is implausible.” Id.
at 19–20.
Google cites Chapman v. New York State Division for Youth for the proposition that a
market definition excluding products or services that plaintiff itself markets is implausibly narrow.
See Def.’s Mot. at 19. But that case is easily distinguished. The plaintiff in Chapman provided a
proprietary crisis management and restraint training program to various entities, including child
care and juvenile facilities. 546 F.3d at 234. The court dismissed plaintiff’s antitrust claims as to
a market for “restraint training services to private child care providers,” because the plaintiff had
failed to show how the restraint training services market to private child care providers was any
different from the restraint training services market to other organizations and businesses, such as
law enforcement agencies, educational facilities, and airlines. Id. at 238. The plaintiff’s
proprietary program was marketed to and used by a variety of entities that were not child care
providers and the “unifying characteristic of this market [was] that each purchaser needs to restrain
individuals, not just children.” Id. That proposed market therefore “clearly [did] not encompass
all interchangeable substitute products even when all factual inferences are granted in plaintiff’s
favor.” Id. (quoting Queen City Pizza, Inc. v. Domino’s Pizza, Inc., 124 F.3d 430, 436 (3d Cir.
1997)).
25
The difference between the proposed market in Chapman and the technologies markets
proposed here is that the plaintiff in Chapman did not offer any reason to think that its individualrestraint program could not be substituted for any other individual-restraint program by any other
purchaser. Here, it is not “facially implausible,” Def.’s Mot. at 19, for example, that the technology
underlying Google Assistant on an Android device is not interchangeable with the technology
underlying Siri on an iPhone or that there is low cross-elasticity of demand between technology
supporting “Hey Google” to activate and operate a phone or tablet and that to activate and operate
a smart TV or doorbell because of the hardware and software disparities and differences in
consumer demand across operating systems and device types. See, e.g., Microsoft Corp., 253 F.3d
at 52 (affirming the district court’s conclusion that “the licensing of all Intel-compatible PC
operating systems worldwide” was a properly defined market even though it excluded “non-Intel
compatible operating systems (primarily Apple’s Macintosh operating system Mac OS) [and]
operating systems for non-PC devices (such as handheld computers and portal websites),” because
the district court had found that consumer demand and hardware differences across operating
systems and device types did not make the products substitutable). Whether these technologies
are in fact interchangeable across operating systems or device types is precisely the kind of
question that should not be resolved at a motion to dismiss. See Eastman Kodak Co. v. Image
Tech. Servs., Inc., 504 U.S. 451, 482 (1992) (market definition is a “factual inquiry”); Todd, 275
F.3d at 199–200 (“Because market definition is a deeply fact-intensive inquiry, courts hesitate to
grant motions to dismiss for failure to plead a relevant product market.”). For now, it is plausible
that the wake word and voice assistant technologies on Android smartphones and tablets are not
“roughly equivalent” to those on non-Android devices other than smartphones and tablets.
See Queen City, 124 F.3d at 437.
26
Finally, Google argues that Sensory does not have antitrust standing in these markets.
According to Google, Sensory alleges no more than “mere temporal proximity” between the
distribution agreements and Sensory’s decline in revenue and only “that Google’s agreements
‘required use of Google’s [wake words] and the Google Assistant,’ not that Google imposed any
requirement on device manufacturers regarding the technology they used in connection with wake
words, voice assistants, and ‘voice search’ on Android phones or tablets.” Def.’s Mot. at 21. Even
if Sensory did compete in the underlying technologies markets, Google argues, it was at most only
“tangentially affected” by “market ripples” caused by the MADAs and RSAs “with respect to the
downstream markets,” which is “not enough to grant antitrust standing.” Id.
The court disagrees. Sensory incorporates into its amended complaint trial exhibits from
Google Search suggesting that Google contemplated a close connection between the distribution
agreements and the technologies underlying wake words, voice assistants, and voice commands.
See Am. Compl. ¶¶ 42–52. One internal record identifies that, under the MADAs, Google “get[s]
“Search widget + Assistant gestures & hotword” and, under the RSAs, “[o]ut-of-box Assistant
gesture, hotword and homescreen exclusivity.” Id., Ex. B, ECF No. 18-2, at 904–05. Another
recognizes the RSAs as securing “Assistant . . . hotword exclusivity” and “Assistant: Full
exclusivity.” Id., Ex. F, ECF No. 18-6, at 047. Drawing reasonable inferences in Sensory’s favor,
it is plausible that this exclusivity also includes the technologies making up Google Assistant and
its associated wake words, even if the records do not explicitly reference them. Sensory therefore
alleges more than mere temporal proximity or tangential effects. So dismissing Sensory’s claims
on this ground is not warranted.
To be sure, Sensory’s pleading is far from robust. See supra Section IV.C. But altogether,
it is enough to survive Google’s motion to dismiss. Sensory has plausibly alleged that Google’s
27
exclusive distribution agreements injured Sensory in the four technologies markets—wake word
technology, voice assistant technology, and voice recognition software (used for search or
otherwise)—in which it once competed.
D. Tying and D.C. Code Claims
Sensory next alleges that Google is engaged in an unlawful tying agreement whereby
Google conditions “distribution of Android, Google Cloud, its Google search services, including
its mobile applications (e.g., Google Phone-top Search and the Android Market Client and Chrome
Browser) and its . . . Google Play store” (the tying product) on both (1) “us[ing] Google’s [wake
words] and Google’s Google Assistant” and (2) excluding “other [wake words] or voice assistants,
voice recognition software, and voice search access points” (the tied products). Am. Compl. ¶ 220.
“A tying arrangement is ‘an agreement by a party to sell one product but only on the condition that
the buyer also purchases a different (or tied) product, or at least agrees that he will not purchase
that product from any other supplier.’” Eastman Kodak, 504 U.S. at 461–62. “Such an
arrangement violates § 1 of the Sherman Act if the seller has ‘appreciable economic power’ in the
tying product market and if the arrangement affects a substantial volume of commerce in the tied
market.’” Id. (quoting Fortner Enters., Inc. v. U.S. Steel Corp., 394 U.S. 495, 503 (1969)).
The court agrees with Google that, as to the tied products for which the court has found
Sensory has not adequately pleaded antitrust standing or a market definition, the tying claim must
be dismissed. See Atl. Richfield Co., 495 U.S. at 334; Todd, 275 F.3d at 200. To the extent the
alleged tied products are in the wake word, voice assistant, and access point markets, the tying
claim is dismissed. See supra Section IV.B, C.1–2. The court will, however, permit the tying
claim to advance as to wake word and voice assistant technology and voice recognition software.
See supra Section IV.C.3.
28
But Google further argues that the tying claim must be dismissed in its entirety because
Sensory has failed to define any of the tying product markets, claiming Sensory has only
“‘mentioned’ these markets ‘but did not define them.’” Def.’s Mot. at 24 (cleaned up) (quoting
Coronavirus Rep., 2021 WL 5936910, at *7–8). But these tying product markets are unlike those
alleged in Coronavirus Reporter v. Apple Inc., where the court held that clearly ambiguous market
definitions like “iOS device market,” “US smartphones,” or “market for smartphone enhanced
commerce and information flow (devices and apps) transacted via the national internet backbone”
were not alone sufficient for the court to “discern what is included and what is not” and conduct
an “analysis of cross-elasticity of demand.” 2021 WL 5936910, at *8. Here, the alleged tying
products are discrete, identifiable operating systems or applications. And this court permitted
similar tying and tied products in Fotobom to proceed to no objection from Google about market
definition. See 719 F. Supp. 3d at 51 (permitting GSuite as the tying product and applications
bundled by the MADAs, including Google Play Store, Chrome, Search, and YouTube as the tied
products). No more is needed to “provide sufficient clarity . . . to assess the threshold question of
whether there is a relevant market” for the tying claims. Def.’s Mot. at 24 (quoting Coronavirus
Rep., 2021 WL 5936910, at *7–8).6
Google makes no other arguments why Sensory’s tying claim should otherwise be
dismissed. Sensory alleges that “Google has market power in the tying product markets” because
OEMs and consumers have no “substitutes” for Android and apps like Play Store or Search.
See Am. Compl. ¶¶ 221–222. And the exclusivity with respect to voice assistant and wake word
technologies required by the MADAs and RSAs referenced in the Google Search trial exhibits,
see supra Section IV.C.3, also renders plausible that, but for the coercive effect of the distribution
6
To the extent Sensory alleges “revenue sharing” is a tying product, see Am. Compl. ¶ 219, the court assumes this is error, as it makes only one errant appearance in the amended complaint.
29
agreements, OEMs would have purchased voice assistant and wake word technologies from
Sensory or other competitors. The court therefore is satisfied that Sensory has plausibly alleged a
viable tying claim. See E & L Consulting, Ltd. v. Doman Indus. Ltd., 472 F.3d 23, 32 (2d Cir.
2006) (“[A]n antitrust defendant charged with illegal tying is entitled to some specificity as to the
conduct alleged to be coercive, the customers who would have purchased a product elsewhere but
for the coercion, the particular products sold as a result of the coercion, the anticompetitive effects
in a specified market, and the effect on the business of the plaintiff.”). The surviving tying claim
shall be as to the conditioning of Android, Google Cloud, Google search services, including its
mobile applications (e.g., Google Phone-top Search and the Android Market Client and Chrome
Browser), and Google Play Store on the exclusive use of Google’s wake word or voice assistant
technologies and voice recognition software.
Finally, because D.C. Code §§ 28-4502 and 28-4503 have been interpreted to mirror
Sections 1 and 2 of the Sherman Act, Sensory’s claims as to those provisions may also proceed
coextensive with the monopolization and tying claims permitted under the Sherman Act above.
See, e.g., Serv. Emps. Int’l Union Health & Welfare Fund v. Philip Morris, Inc., 83 F. Supp. 2d 70,
90 n.33 (D.D.C. 1999) (citing D.C. Code § 28-4515), rev’d on other grounds, 249 F. 3d 1068 (D.C.
Cir. 2001); D.C. Code § 28-4515 (“It is the intent of the Council of the District of Columbia that
in construing this chapter, a court of competent jurisdiction may use as a guide interpretations
given by federal courts to comparable antitrust statutes.”).
E. Motion for Leave to File Surreply
Sensory’s motion for leave to file a surreply in response to Google’s reply is denied.
Its proposed surreply is simply a regurgitation of trial exhibits and transcripts from the remedies
phase of Google Search under the guise of presenting “new evidence.” See generally Pl.’s Mot.
30
for Leave to File a Surreply in Opp’n to Def.’s Mot., ECF No. 23, Pl.’s Surreply in Opp’n to Def.’s
Mot., ECF No. 23-1. That is not a valid ground on which to grant leave to file a surreply.
See Neville v. Burrows, No. 22-cv-3246, 2024 WL 578986, at *6 (D.D.C. Feb. 13, 2024) (“To the
extent Plaintiffs are seeking to raise new factual allegations based on this ‘recently discovered new
evidence,’ the Court notes that a surreply . . . is an inappropriate vehicle to raise such new
allegations.” (cleaned up)); see also González-Vera v. Townley, 83 F. Supp. 3d 306, 315 (D.D.C.
2015) (“[T]he point of a surreply is to allow a party to respond to an opponent’s new facts and
arguments—not to unearth facts and arguments never before raised.”).
V. CONCLUSION
For the foregoing reasons, Google’s Motion to Dismiss, ECF No. 20, is granted in part and
denied in part. All three counts of Sensory’s Amended Complaint are dismissed only insofar as
they concern the following markets:
• “general search services,” Am. Compl. ¶ 86;
• “general search text ads,” id.;
• “general search advertising,” id. ¶ 178;
• “access points to general search services on phones and tablets running the Android
operating system,” id. ¶ 226;
• “access points to general search services and voice searching on Android phones
and tablets,” id. ¶ 211;
• “wake words on smartphones and tablets running the Android operating system,”
id. ¶ 152; and
• “voice assistants on smartphones and tablets running the Android operating
system,” id. ¶ 154.
Google shall answer Sensory’s Amended Complaint as to all three counts only insofar as they
concern the following markets:
• “wake word technologies” on Android smartphones and tablets, id. ¶ 140;
• “voice assistant technology on smartphones and tablets running the Google Android
operating system,” id. ¶ 141.
• “voice recognition software (speech to text)” on Android smartphones and tablets,
id. ¶ 167; and
31
• “voice recognition software (speech to text) used for search” on Android
smartphones and tablets, id. ¶ 228.
Finally, Sensory’s Motion for Leave to File a Surreply in Opposition to Defendant Google LLC’s
Motion to Dismiss, ECF No. 23, is denied.
Dated: July 13, 2026 Amit P. Mehta
United States District Judge
32