Sumanyu Sharma (Hamming AI) discusses reliability failures in voice AI agents. Drawing on his prior work listening to thousands of hours of police radio, he contrasts locally-occurring crime with centrally-scaled voice agents, where a single change propagates to all users. With roughly a trillion phone calls per year and observed error rates near ten percent across ~10,000 monitored agents, failures include agents skipping eligibility checks, applying unauthorized discounts, and falsely confirming bookings. He recommends an iterative testing loop — find failures, size them by frequency and severity, fix, verify the fix, and keep monitoring — emphasizing manual call review as a starting point and cross-conversation analysis for scale. He also reports that adversarial testing breaks roughly one agent in five, and covers attacker scenarios where bad actors dial voice agents. (Talk title also references "1 trillion phone calls/yr, 10% error rate.") # Channel Presentation Observations — Hamming AI (Sumanyu Sharma Talk) ## Explicit Self-Description (Creator Claims) - **Founder/CEO identity**: Speaker introduces himself as "Suman Yu [Sumanyu Sharma] and I'm the founder and CEO of Hamming," working on "voice agent reliability and safety." - **Company product claims**: Hamming "shipped a red teaming product back in April" and claims "we can probably break one in five agents at this point," tested "across financial services, healthcare, consumer." - **Operational claim**: "In practice, we currently monitor 10,000 agents and the error rate is closer to 10% in practice." - **Business solicitation**: The talk closes with an explicit pitch — "if you are deploying voice agents and want to validate whether your architecture or your eval are set up correctly, please come and talk to us." ## Intended Audience - **Voice AI builders and deploying teams**: Frequent direct address to teams operating voice agents — "most teams will typically start by listening to calls manually," "some of the best teams that we work with." - **Technical practitioners in evals/reliability**: References to "LLM as a judge," "AB testing," "pre-deployment synthetic testing," and "cross-conversation analysis" suggest an engineering audience. - **Startup/tech event attendees**: Audience interaction ("Anybody here use Citizen app? Awesome. Thank you.") and references to YC figures ("Gary Tan and Polygram") indicate a startup-conference setting. - **Potential hires and customers**: "If you are interested in working in this space, please come and talk to us." ## Claimed Scope of the Channel/Company - **Voice agent testing, evaluation, and monitoring**: Demonstrated via a full reliability framework (identify → prioritize → fix → execute → verify → monitor) and discussion of per-call scoring, manual evals, and cross-call analysis. - **Adversarial/red-team testing**: "Our new recommendation now is to run 24/7 red teaming for your agents." - **Cross-industry applicability**: Financial services (credit card freeze example), healthcare (appointment scheduling failure), consumer (drive-thru ordering example). ## Demonstrated Presentation Patterns (Observable) - **Personal anecdote as evidence**: A failed physician appointment booking ("I showed up to the appointment and turns out I was not actually on the schedule"). - **Quantified framing**: "At least a trillion calls that are done every single year"; "If you assume a 1% error rate that is still 10 billion incidents per year." - **Analogy-driven structure**: Extended comparison of crime (from his Citizen background) to voice agent failures — "trash fires" vs. safety risks, hyper-local vs. centralized "blast radius." - **Cautionary/urgency tone**: "Voice agents scare me more because they're finally graduating from demos and PCs to production... I'm personally nervous." - **Embedded third-party media**: Plays a Twitter-sourced audio clip of a confused trade-in call as a failure example. ## Distinguishing Claims from Demonstrations - **Claims (unverified creator statements)**: 10,000 monitored agents; 10% observed error rate; one-in-five red-team break rate; trillion calls/yr statistic. - **Demonstrated**: Domain expertise in voice AI failure modes, evaluation methodology, and adversarial testing; a sales-oriented closing; structured framework thinking borrowed from named sources ("a couple of my friends who were OG growth folks at Facebook"). # Delivery Observation: Sumanyu Sharma — Voice AI Reliability Talk ## Overall Tone - **Conversational, founder-style pitch** delivered with casual, spoken-language markers throughout ("you know," "uh," "um," "right?"). - Mixes **personal nervousness with enthusiasm**: "We should be super excited, but I'm nervous. I'm personally nervous." - Uses **humor and self-deprecation** to keep the audience engaged: "So it's time for me to shill." ## Recurring Phrases & Verbal Habits - **"You know"** appears extremely frequently as a filler/connective phrase (e.g., "It's getting much faster to build products and voice experiences... in a pretty short period of time"). - **Repeated intensifiers**: "super, super good," "super super high," "pretty pretty critical," "quite quite massive." Doubling intensifiers is a stable pattern across the talk. - **Restatement for emphasis** — repeats key ideas immediately after stating them: > "This is still the number one problem." > "This is real. This is real time." - Frequent rhetorical questions to pivot sections: "So the natural question is how do you make these incidents much more visible and obvious?" and "Now, what happens when mythos learns how to dial?" ## Teaching Habits & Structure - **Analogy-driven teaching**: The core structure contrasts crime (from his Citizen background) with voice agent failures — "Now, let's compare crime to voice agents" — sustained across most of the talk. - **Numbered framework delivery**: Presents a step-by-step process explicitly ("So step one is to identify... Step two is to prioritize... Step three... Step four execute. Step five..."). - **Concrete anecdote before abstraction**: Opens with personal stories (Citizen alerts, the ice-cream thief, his missed physician appointment) before introducing statistics and frameworks. - **Audience interaction**: Directly polls the room — "Anybody here use Citizen app? Awesome. Thank you." ## Vocabulary & Reference Style - **Startup/tech ecosystem vocabulary**: "OG growth folks at Facebook," "P 0 target areas," "bake off," "AB testing," "LLM as a judge" (transcribed as "element as a judge"), "blast radius." - **Named product drops** used as credibility/authenticity signals: "Aqua voice, Super Whisper, uh, Whisper Flow," plus references to "Gary Tan and Polygram." - **Concrete quantitative anchors** for claims: "at least a trillion calls," "10 billion incidents per year," "the error rate is closer to 10%," "we can probably break one in five agents." ## Personal Framing & Vulnerability - Repeatedly uses **his own failed appointment as a recurring case study**: "just like it happened for me," "It already did for me." - Escalation via **hypothetical questions** to dramatize stakes: "What if this was actually your parent? What if this was your grandparent?" ## Closing Pattern - Ends with a **direct recruitment/sales call to action**, framed jokingly: "it's time for me to shill... please come and talk to us... And here's my number. Here's my WhatsApp." ## Uncertainty Notes - Fillers and some phrases appear garbled in the transcript (e.g., "the long tail is still Hey, Gorov," "mythos learns how to dial" — likely a model name), so exact wording in a few spots is uncertain. - Some doubled-word patterns may reflect transcription stutter ("experiment experimenting") rather than delivery habit; I've excluded those from the pattern claims. # Observation Document: Voice AI Reliability Crisis — Sumanyu Sharma (Hamming AI) ## Speaker - **Sumanyu Sharma** (transcribed as "Suman Yu"), founder and CEO of **Hamming AI**. Prior background at **Citizen** (NYC), where he worked analyzing police radio data and sending alerts. ## Background & Framing - At Citizen, he says they listened to "thousands of hours" of police radio data and sent millions of alerts across cities including SF, NY, LA, Chicago, Baltimore. - Claims voice agents "scare me more" than crime because they are "graduating from demos and PCs to production." - **Personal qualification:** "We should be super excited, but I'm nervous. I'm personally nervous." ## State of Voice AI (as expressed) - Early 2024: voice was "just starting to work"; products like Aqua Voice, Super Whisper, Whisper Flow are now "super, super good" — he jokes you'd have to pay him a lot to stop using them. - Improvement drivers: better underlying infrastructure, better orchestration layers; ~60% good experiences buildable quickly. - **Qualification:** "the long tail is still... wide away"; speech models improving, teams experimenting with hybrid voice-to-voice and cascading stacks. - Agents now connected to calendars, CRMs, HR systems, reservation systems — can take actions. ## Core Claim: Reliability Is the #1 Problem > "Reliability is still the number one problem holding back most voice agent deployments at scale." ### Evidence cited - **Twitter example (trade-in confusion):** agent gave wrong info; a person ("Alex") had to call back to fix the loss of trust. - **Personal example:** booked a physician appointment via an agent; showed up but "was not actually on the schedule." Wasted 2 hours. Escalation framing: "What if this was your parent? Your grandparent? What if it was a procedure?" ### Scale and error rate - At least **1 trillion calls/year**; majority expected to be conversational voice agents within ~5 years. - Even at 1% error rate: **10 billion incidents/year**. - **In practice:** monitoring ~10,000 agents, error rate "closer to 10%." - Failure examples listed: skipping eligibility/verification steps while claiming to find the right policy; applying unapproved discounts; mishearing users; incorrect information; claiming appointments booked when not. ## Crime vs. Voice Agents Comparison - **Trend:** crime decreasing (he hopes it crosses the x-axis); voice agent deployment rising fast. - **Severity spectrum (crime):** from "trash fires" (funny/annoying) to mass shootings. - Voice equivalents: repetition/misunderstanding → drive-thru scenario (Taco Bell/McDonald's) where a vegan order or **peanut allergy** mishandled is "definitely a safety concern at scale." - **Locality:** crime is hyper-local, decentralized; voice agents are **centralized** — a single prompt or architecture change affects millions of users; "the blast radius is quite massive." ## Proposed Framework (borrowed from "OG growth folks at Facebook") 1. **Identify** challenges in the conversation experience 2. **Prioritize** by impact size (frequency × severity) 3. **Understand** how to fix 4. **Execute** 5. **Check** the change worked / no regressions 6. **Monitor** in production ### Identification quadrant - **Known problems** (turnover latency, interruptions, ASR issues) — track over time. - **Emerging patterns** — only visible across many conversations. - **Coverage axis:** manual listening is the best starting point ("I don't think you should skip that step") — builds intuition but doesn't scale; teams use 5–10 rubric spreadsheets; then eval products (LLM-as-judge, deterministic/stochastic scoring) — but these check known problems, not novel insights. Hamming focuses on **cross-conversation analysis**. ### Prioritization - **One-off, low impact:** "who cares?" - **Low impact, systematic:** still annoying at scale; worth solving in bake-offs. - **One-off, high impact:** "hope it doesn't happen." - **Systematic, high impact:** P0. Example: finserv agent freezing credit cards — failure is a "massive fail." ### Testing fixes - **Naive:** replay the exact failing call 5–50 times, measure pass probability. - **Better:** keep intent, vary wordings, patterns, accents, styles, add intents. - **A/B testing** needed for hypotheses hard to test synthetically — e.g., first 5 seconds of outbound calls (vocal quality, word choice): "You can't really do it through simulations alone." ## Adversarial Threats - Question: "What happens when [models] learn how to dial?" — if models can extract NSA trade secrets, they can "seduce you into revealing PHI and PII." - Both **voice agents and humans** will be targeted: capability pressure expands attack surface ("the more the capability, the bigger the surface area"); more natural-sounding agents make humans easier to trick. - **Red teaming results (claimed):** product shipped April; "we can probably break one in five agents" across financial services, healthcare, consumer. Bypassed verification; prompt-injected several agents and "gotten data we should not have." - "This is not theoretical." ## Recommended Defenses 1. **Deep pre-deployment testing** (text-to-text or voice-to-voice, each with pros/cons). 2. **Great monitoring**: per-call scoring, manual evals, cross-call analysis — monitors both agent behavior and whether users are adversarial. 3. **New recommendation:** 24/7 red teaming, especially when the cost of bad interactions is large. ## Closing Contrast - Voice agents have potential to make the world "feel much more human" vs. IVR trees, chatbots, hold queues. - Crime "happens to somebody else"; with voice agent incidents, "one of these incidents is going to impact you. It already did for me." ## Note on Evidence Status - Claims about the ~10% error rate, 10,000 monitored agents, and one-in-five red-team break rate are **speaker-asserted figures** with no cited methodology in the transcript. # Speaker & Provenance Record — *1 Trillion Phone Calls/yr, 10% Error rate: The Crisis in Voice AI* ## Speaker Identification | Speaker | Role | Evidence | |---|---|---| | **Sumanyu Sharma** (rendered "Suman Yu" in transcript) | Founder & CEO of **Hamming AI**; presenter of this talk | Self-introduction: *"My name is Suman Yu and I'm the founder and CEO of Hamming"* | - No additional named speakers. The video includes short **unattributed third-party voice clips** (from a trade-in call example) and background **music** — the clip participants are not identified. - Prior background stated by speaker: worked at **Citizen** (New York), monitoring police radio data for real-time alerts. - No curated metadata beyond the title was supplied; speaker name normalized from title metadata ("Sumanyu Sharma, Hamming AI") where the transcript differs. ## Referenced Sources & Third-Party Claims All claims below originate from the speaker unless noted: - **Twitter example** (found "two weeks ago"): a customer ("Alex") confused by a voice agent during a vehicle trade-in inquiry, with audio of the exchange played. - **Personal anecdote**: speaker's physician appointment failed to be scheduled by a voice agent, resulting in being turned away and losing 2 hours. - **Companies/products named**: Citizen, Aqua Voice, Super Whisper, Whisper Flow, Taco Bell, McDonald's, Facebook (source of a growth framework from friends who were "OG growth folks" there), Gary Tan and Y Combinator (implied via "Polygram" [unclear in transcript]). - **Quantitative claims** (unverified by transcript): at least **1 trillion calls/year**; ~**10% observed error rate** in practice across **10,000 monitored agents**; red-teaming product can break roughly **1 in 5 agents**. - **Promotional segment** (explicit): speaker states *"it's time for me to shill"* — invites audience to talk about **hiring at Hamming** ("we burn a lot of tokens") and to validate voice agent architecture/evals; shares his phone number and WhatsApp. ## Promotion Inventory - **Hiring pitch** for Hamming AI — flagged per speaker's own framing as a "shill." - **Product mentions**: Hamming's monitoring/cross-conversation analysis work; a **red-teaming product** shipped in April (framed as evidence, but is the speaker's own company's offering). ## Availability & Evidence Channels - **No comments or follow-up channel** recorded in the source material beyond the in-person invitation to find the speaker outside and contact via shared WhatsApp number. - Slides/frameworks referenced but not supplied in transcript text (e.g., the two-axis monitoring matrix: known vs. emerging problems; coverage axis). ## Notes for Later Authors - The **transcript is auto-generated** with errors (e.g., "Suman Yu," "Gorov," "mythos," "promject," "stoastic") — treat quoted strings cautiously. - Speaker's claims about industry-scale error rates and red-team success rates are **self-reported and uncorroborated** within the source. My name is Suman Yu and I'm the founder and CEO of Hamming. And before working on voice agent reliability and safety, I worked at a company called Citizen out of New York. Anybody here use Citizen app? Awesome. Thank you. Uh, and at Citizen, we listened to crime, thousands of hours of police radio station data and sent millions of alerts to users in San Francisco, New York, LA, uh, Chicago, Baltimore, and so on. Some obviously gory and pretty sad. Uh, but others more funny like a person stealing bags of ice cream from Safeway or report of a man hanging off the side of the house after a woman stole his ladder. If I actually take a look at the citizen app right now for those who are customers or users, I can see that there is a man yelling at person. There's indecent exposure. This is real. This is real time. This is, you know, couple hours ago. These are real-time alerts that we're sending. Now, voice agents scare me more because they're finally graduating from demos and PC's to production. We should be super excited, but I'm nervous. I'm personally nervous. Uh, they're talking to users at a scale that would make Gary Tan and Polygram proud. When I got started in voice agent reliability in early 2024, voice was just starting to work. It was not quite good yet, but it was just starting to work. You would have to pay me a lot of money for me to stop using, you know, Aqua voice, Super Whisper, uh, Whisper Flow, and so on. These products are just getting super, super good. And a big reason is because the underlying infrastructure is getting better, and the orchestration layer is getting meaningfully better. It's getting much faster to build products and voice experiences that maybe are 60% good in a pretty short period of time, but the long tail is still Hey, Gorov. the long tail is still uh wise away. I think speech speech models are getting better. Um teams are experiment experimenting with hybrid architectures of combining more voicetovoice modalities and also cascading stacks to make the experience reliable but still pretty low latency. Things are obviously getting better. Agents are being connected to calendars, CRM, HRs, reservation systems, and so on. Voice agents can now take actions. However, reliability is still the number one problem holding back most voice agent deployments at scale. This is still the number one problem. This is an example I found on Twitter pretty randomly, you know, two weeks ago and a person is trying to get information for a tradein and gets absolutely confused with the information that they're receiving. Alex now has to correct for this loss of trust but trying to, you know, call the person and see what see what happened and fix the situation. Let me see if audio works here. \>\> Screwed up with another customer. We're getting it fixed, but I got to call him and see if I can work it out. \>\> I'm like, dude, half the time I'm like, I don't know if I'm talking to AI. I don't know if I'm talking to a person. It was just confusing, but we got there. \>\> It probably is AI and human. And \>\> so I think voices sound very confident. They sound very natural, but the information provided is often, you know, not correct. That's the biggest problem here. This example is more personal. I had booked an appointment with a physician a couple weeks ago or I thought I did. I showed up to the appointment and turns out I was not actually on the schedule. So the front desk, you me turned me away. I wasted 2 hours. For me, this was a waste of time. But what if this was actually your parent? What if this was your grandparent? What if this appointment was for a procedure instead of a regular checkup? The costs for these different permutations of the same failure mode can actually be super super high. Now, let's compare crime to voice agents. Um, I think observation number one is crime is actually decreasing over time. This is a good thing and I hope it crosses the x- axis at some point you know in the future. Voice on the other hand is generally taking off right we're seeing a pretty fast takeoff of voice agents being deployed in production. There's at least a trillion calls that are done every single year and majority of these will be done by conversational voice agents over the next you know five years. If you assume a 1% error rate that is still 10 billion incidents per year. That's a lot. In practice, we currently monitor 10,000 agents and the error rate is closer to 10% in practice. These range from agents saying they found the right policy when they actually skipped the eligibility or verification steps or applying discounts when they were not really supposed to, misharing what the person said, providing incorrect information, or claiming they booked an appointment when they actually did not, just like it happened for me. Now, not every single call has an equally, you know, bad cost. Uh, some range, you know, in the crime land, some range from trash fires, which are kind of funny, annoying, not really hurting somebody. For a voice equivalent, that would be annoyances like repetition, um, or just sort of not quite understanding what the user is saying. all the way to safety risks like mass shootings or in the voice agent equivalent, it would be um a drive-thru that's deploying um voice agents at scale like a Taco Bell or McDonald's and a person orders a vegan burger with peanut allergies. If one of those two situations are not handled correctly, that is definitely a safety concern at scale. The other big difference between crime and and voice agent deployments is is crime generally tends to be pretty hyper local, tends to be very decentralized, right? Things like robbery or motor vehicle theft or lararseny. They're impacting a finite set of individuals that are involved in that um situation. On the other hand, voice agents are much more centralized. a single prompt change or an architecture change can have pretty massive implications downstream for all of the millions of you know users that are um in the crossfire. So the blast radius is is quite quite massive. So the natural question is how do you make these incidents much more visible and obvious? That's the kind of obvious question here. I'll borrow a framework from a couple of my friends who were OG growth folks at Facebook. So step one is to identify okay what are all the challenges and problems that um exist in your conversation experience. Step two is to prioritize an impact size. There's a frequency and severity analysis that's pretty important. Step three is to understand okay how do we actually fix this? Step four execute. Step five okay did my change actually work and did it cause any regressions somewhere else. And lastly we continue to monitor in production. On the y- axis, I think it's important to highlight there are known problems that already exist. Things like turnover latency, interruptions, um maybe some ASR problems you're, you know, aware of. And these are known problems that exist that the team should track over time. On the other axis is actually emerging behavior or patterns that are only obvious across lots of conversations. Um on the x- axis, you have coverage just like insurance. Are you analyzing few conversations? Are you analyzing many, many conversations? Most teams will typically start by listening to calls manually. And I think that's the best place to start. I don't think you should skip that step. There's a lot of depth and insights to get by actually listening to specific conversations and building that texture that that comes from that intuition. However, it's obviously not scalable. So most teams end up having a spreadsheet of I don't know five or 10 different rubrics around greetings, closing, validation, um, core logic and so on. To scale that up even further, you then end up investing in some eval product, right? You might run some element as a judge and compute classic metrics and also more more deterministic and stoastic scoring logic. Um, but there you're still stuck with checking for consistency of known problems, but you're not really discovering novel insights that are actually happening across conversations. We're spending a ton of time on performing cross conversation analysis, not a pattern on a single call, but across conversations. And some of the best teams that we work with are are doing the same. Now, to prioritize an impact size, I think there's problems that are one-off that are low impact. I mean, who cares? uh even low impact and systematic problems in the crime world that would be a trash fire in a voice aation world it could be some repetitions the team is experiencing they're still annoying at scale and if you are doing a bake off it's still worth solving for them I would not ignore these class of problems oneoff and high impact well hope it doesn't chronic and I think systematic and high impact are obviously the P 0 you know target areas um for the team to solve an example of that would be in a fins serve capacity There's a voice agent that um helps users freeze their credit cards. And if it doesn't do that, well, that's a massive fail. All right. So, understand and execute. I'm pretty sure everyone's doing this. Please fix my agent. Uh I think fixing or rather attempting to make a fix is the simplest and the lowest effort component of this debugging pipeline and loop. Um the next step is all right, I made a change to my system. How do I actually know this thing works um for real? A great way that's naive is to take a real call, for example, in my case, I booked an appointment and it didn't get scheduled and replay that exact conversation and run that maybe 5, 10, 20, 50 times and see, okay, what is my probability of passing this type of issue? A better way is to keep the same intent but change the wordings, change the patterns, change the accents, change the style, add one more intent to the mix. And that gives teams much more, you know, better coverage to feel confident that yes, I actually made a change and my changes are net positive instead of net negative. There are certain fixes and I guess hypothesis that are very difficult to test in a pre-eployment synthetic setting. And so AB testing ends up being, you know, pretty pretty critical for those circumstances. For example, if you have an outbound agent, the first 5 seconds of a conversation tends to be the most important. And so the vocal quality um and the specific words you end up using, they matter the most. And so AB testing that is the only way in in kind of real life setting to to get results. You can't really do it through simulations alone. And so there we have the loop. Identify, prioritize, impact size, understand the fix, execute, check, make sure it didn't break anything, and then continue monitoring. So I think making voice agents useful is already hard as it is. even when dealing with earnest users on the other line, right? These are people who who just want their problem solved. They're not trying to mess with you. These are like legit normal people. Now, what happens when mythos learns how to dial? So, if it can extract trade secrets and uh you know, from the NSA, it can certainly, you know, seduce you into revealing PHI and PII data as well. And I think both voice agents and humans will \[clears throat\] be targeted here. Voice agents because there's a pressure to make these more capable. Give them access to more data. Give them access to more tools. Deploy them quickly. The more the capability, the bigger the surface area. This is this is pretty pretty common sense. And the more the voice agents become natural and human sounding, the more humans will be tricked along the way as well for those who are weaponizing. Uh we ship a we shipped a red tipping product um back in April just to test out this hypothesis for how many agents can we actually break from a adversarial capacity and we can probably break one in five agents at this point. We've tested this across financial services, healthcare, um consumer and so on. We've bypassed verification. Uh we've definitely had agents, you know, we've been able to promject uh several agents and and gotten data we should not have. So this is not theoretical. This is actually a real a real concern. I think the only real defense against the dark arts is step one to invest deeply in pre-eployment testing. This could be textto text. This could be voice to voice. There's pros and cons to both. Happy to chat offline if folks are interested. And this is just making sure you're not self-owning, you know, when you're talking to real people who just want to get their problem solved. Step two is to have a great monitoring system of all kinds. And I've highlighted, you know, different flavors of monitoring per call scoring, manual kind of evals, you know, listening to conversations and cross call analysis. And this is helpful both for monitoring what the agent is saying and behaving and how it's actually doing, but also the users. Are the users being adversarial? Are they being annoying? Are are they trying to trick the agent into doing things it's not supposed to be doing? And I think our new recommendation now is to run 24/7 red teaming um for your agents, especially if you believe the cost of bad interactions can be can be rather large. So, I think voice agents um have this awesome potential of of making the world feel much more human compared to interacting with clunky IVR trees or chat bots or worse um being stuck on a on a hold. And when we think about crime, we often think of crime happening to somebody else. You know, crime does not happen to you typically with voice agents, especially bad actors. As these agents are deployed and as bad actors start to exploit a lot of the vulnerabilities, the number of incidents is about to kind of go way way up. And so the reason I fear voice agents more than crime is that one of these incidents is going to impact you. It already did for me. Awesome. So it's time for me to shill. Well, we burn a lot of tokens. If you are interested in working in this space, please come and talk to us. And if you are deploying voice agents and want to validate whether your architectured or your eval are set up correctly, please come and talk to us. We'll be outside. And here's here's my number. Here's my WhatsApp. Thanks everyone. \>\> \[music\]