I think they’re vastly understating/oversimplifying how they use WASM. For audio analysis and operations on audio binary/encoded data, it’s quite possible that WASM is a very good fit. Maybe offloading that work to a JS worker would be sufficient—you’re right that JS VMs are extremely fast—but I wouldn’t necessarily discount this WASM use case either. And I’m saying that having explored these kinds of optimizations enough to generally recommend against using WASM unless you have strong reason to believe you’d benefit from it (ideally with benchmarks to prove it).
I’d also caution that using Web Workers isn’t always so obvious either (and the same applies to server side threading eg on Node). There’s significant overhead (runtime) in spinning up a worker, and in every communication between threads—enough to negate their benefit for many use cases. Both WASM and workers have roughly the same perf downsides, and both should generally be justified by real measurement of their impact.
I’d also caution that using Web Workers isn’t always so obvious either (and the same applies to server side threading eg on Node). There’s significant overhead (runtime) in spinning up a worker, and in every communication between threads—enough to negate their benefit for many use cases. Both WASM and workers have roughly the same perf downsides, and both should generally be justified by real measurement of their impact.