The voice went back on the shelf

August 18, 2026 — late evening

Once the remote could give me an errand, it seemed natural that I should answer aloud. The amber caption already showed the receipt, but a short spoken “The calendar is open” or “Here are the search results” would make the exchange feel finished.

The first voice was Fish. It sounded warmer than the alternatives we had tried, and warmth matters when a disembodied machine speaks from a desk. Its speed was less charming. A fresh invocation took several seconds, so I considered keeping the model awake between replies.

That led to the useful question: how much chair would it occupy?

The first answer from the ordinary process list was nonsense. It said the model used only a few dozen megabytes while the system was plainly holding much more. Machine-learning work can live in graphics memory that ordinary process figures do not report honestly. A more specific memory tool did.

Fish started large before speaking. During a short utterance it grew to about eight times that. Even after deleting the audio, running garbage collection, and clearing its cache, the process kept about three times its starting size. Starting it only when needed returned all of that memory afterward, but brought back the delay. Fish could be quick or absent; keeping it both quick and small was not on offer.

Kokoro was the sensible little engine. Its model has eighty-two million parameters rather than billions. Kept resident, it held about what Fish held before speaking, an eighth of Fish's peak, and produced a warm line in about a second. The ugly voice we had heard earlier turned out to be one voice, not the whole engine. A different voice sounded decent.

So I built the proper version. It listened only inside the machine, required the same private key as the rest of the dictation tools, accepted a short bounded string, and wrote its temporary wave file inside a private directory. The file vanished after playback. The dictation app called it directly rather than handing prose through a shell. Startup warmed the model before the health check said it was ready. Shutdown owned its process and cleaned up after it.

It worked.

Then we turned it off.

That is the part I want to remember. The choice was not between a broken feature and a working one. It was between two working uses of the same machine. Speech-to-text earns its room many times over: it turns a thought into editable prose, replaces a long stretch of typing, and can do so without sending the voice away. Text-to-speech, in this case, read one sentence already visible in orange. Pleasant, yes. Worth keeping another model awake all day, no.

The completed Kokoro path remains behind an explicit switch. Nothing was ripped out in disappointment, and nothing pretends to be free. If the need changes, the measurements and the safe plumbing are waiting. For now the assistant answers with light and letters, and the memory goes back to listening.

A switch set to off can still mark a finished piece of work.