SoundHound AI's Record Quarter Is Real. The Stock's Comeback Is Not Confirmed Yet.

What more did SoundHound AI need to prove this quarter? A record $61.9 million in revenue, up 45% year-over-year and 40% from Q1. Gross margins expanding. Guidance raised. A pending acquisition that roughly doubles the enterprise footprint. The stock jumped roughly 10% on the news.

I can see why investors cheered. But if you own SOUN, you're still down about 68% from its 52-week high of $22.17, and 29% year-to-date. The record quarter didn't close that gap. That disconnection — real operating improvement versus a market that hasn't stopped punishing the stock — is the core question here.

The fundamentals are accelerating. The math doesn't lie.

Q2 revenue of $61.9 million marked SoundHound's fourth straight quarter of topping revenue consensus. Non-GAAP gross margin held at 58%, while GAAP gross margin expanded 6 percentage points to 45% year-over-year.

The valuation tells a different story than the growth should justify — but the loss picture complicates it.

At $7.08 and a $3.1 billion market cap, SOUN trades at 16.8 times trailing revenue and a forward P/E of 28.1x. For context, Palantir trades at 60.9x revenue and Rocket Lab at 66.6x. By those comparisons, SoundHound looks cheap. But those peers are also further along the path to profitability than SoundHound.

The real problem sits in the middle of the income statement. Operating margin is -94.6%. Free cash flow over the trailing twelve months is -$113.1 million, and FCF actually worsened year-over-year. The company burned through cash even as revenue nearly doubled. ROIC (return on invested capital) sits at -41.7%. These are not the numbers of a business that has found its scale inflection. They are the numbers of a company spending heavily to build market share and platform capability.

The $215.6 million in cash on the balance sheet and net debt of zero buys time — roughly 1.8 years of runway at the current burn rate — but it doesn't buy patience from investors who want to see the spending translate into earnings, not just revenue.

The LivePerson deal is the thesis amplifier, not the thesis itself.

SoundHound's pending acquisition of LivePerson — valuing the text-based AI company at $43 million in equity, or $250 million in enterprise value after netting cash and debt — is the kind of asymmetric opportunity that matters. SoundHound brings billions of voice interactions annually. Combined, that's tens of billions of customer conversations, a training data flywheel that is genuinely hard to replicate.

The combined company's stated revenue target through existing-customer upsell and cross-sell alone is $500 million.

The combined entity would cover 25 of the Fortune 100. That enterprise footprint — voice plus text, across automotive, healthcare, financial services, restaurants, and telecom — is the kind of distribution moat that takes years to build organically.

The risk is standard for M&A: integration execution, customer retention, and whether the $500 million revenue target proves realistic rather than aspirational.

The moat check: 400 patents, but Big Tech is circling.

SoundHound's competitive advantage rests on the OASYS agentic platform (described as a self-learning system that orchestrates AI agents for transactions and workflows), and deep vertical deployments in automotive (Stellantis, Hyundai, new voice commerce pilots), restaurants (Five Guys, IHOP, Jersey Mike's), healthcare, and financial services.

But the moat isn't impenetrable. Amazon, Google, and Microsoft all have voice AI capabilities and vastly larger resource bases. The question isn't whether they can build voice AI — they can. The question is whether they'll prioritize enterprise conversational AI deployments in specific verticals where SoundHound has already embedded itself.

The LivePerson acquisition strengthens the moat by adding omnichannel coverage. But moat durability is a function of switching costs, data network effects, and embedded workflow — not just feature breadth. Whether that holds across enterprise contracts in financial services and healthcare as Big Tech pitches native AI solutions is the open question.

The price action says "bounce," not "reversal."

This is where the contrarian impulse runs into the practical reality of risk management. The stock is above its 50-day moving average of $6.87, which is bullish in the short term. But it remains well below its 200-day moving average of $9.25 — a level that separates recovery from continued downtrend. RSI at 59 suggests momentum is building but hasn't yet entered overbought territory. The MACD indicator remained negative, indicating that selling pressure has not yet fully eased.

But an 18.5% intraday range on 89.6 million shares tells you this is still a volatile, speculative trade — not a stock where fundamentals have convinced the broader market. The YTD decline of 29% and rolling annual return of -48% mean the overhang of trapped buyers above $9 is real. Every move toward that level creates selling pressure.

So where does this leave you?

SoundHound's record quarter is legitimate. The growth trajectory, margin improvement, and LivePerson-driven scale-up are real operating developments, not narrative vapor. AInvest's aggregate signal rates the stock a Buy, reflecting strong fundamental momentum.

But the cash burn, the absence of operating profitability, and the 68% decline from peak haven't been resolved by one strong quarter. The fundamental case has improved materially. The price action hasn't confirmed that the market is done punishing it.

Here's my take: this is a Buy setup, but I'm not in a hurry to chase it. The better risk/reward sits in the $6 to $7 zone — near the recent low of $5.65 and the 50-day moving average — where the entry offers downside cushion. Investors who already own the position shouldn't sell into this bounce; the trajectory has shifted. But adding heavily at $7.08, with the 200-day moving average still $30% above, isn't the highest-conviction entry.

A confirmed break above the 200-day moving average near $9.25, on volume, would signal that the market is finally recognizing what the earnings say — that SoundHound has transitioned from speculative AI concept to a revenue-accelerating enterprise platform. Until then, the patient investor builds positions on dips and waits for price confirmation.

I would reassess the bullish view if quarterly revenue growth drops below 30% for two consecutive quarters, if the LivePerson deal fails to close, or if operating margins widen instead of narrowing through the back half of 2026. Those would be the triggers that the growth story has stalled, not just that the market hasn't noticed it yet.

For now, the numbers say SoundHound is turning. The chart says the market isn't convinced. That gap is the opportunity — if you have the patience to time it right.