Model Issues — Speakly

Local models are downloaded from Hugging Face and kept in Speakly's application data folder. You manage them in Settings → Transcription, under Local Model (On-Device).

Where models are stored

macOS: ~/Library/Application Support/Speakly/whisper-models
Windows: %APPDATA%\Speakly\whisper-models

One .bin file per model — for example ggml-large-v3-turbo-q8_0.bin. A download in progress is written to that same name plus .tmp and only renamed once it finishes, so a half-downloaded model can never be mistaken for a complete one.

A download failed

Speakly shows Download failed. Please try again. and deletes its own .tmp file. To retry:

Check you are online and have the disk space — the large-v3 files run to gigabytes.
Open Settings → Transcription and click Download on the model again.
If a stray *.bin.tmp file is still sitting in the folder above — possible after a crash or a force quit — delete it, then download again.
If the model is labelled Downloaded but still fails, click its trash icon, confirm Delete Model?, and download it once more.

Cancelling is safe

Clicking Cancel during a download stops it and removes the partial file. Nothing is left half-written under the model's real name, so the next Download starts clean.

What the error messages mean

Model not found, in the recording pill — whisper.cpp could not initialise the model file. Delete the model in Settings → Transcription and download it again.
Local transcription failed, in the recording pill — every other local failure, including the missing-model case whose full text is Model "large-v3-turbo-q8_0" is not downloaded. Please download it in Settings > Transcription.
Local transcription returned empty text — the model ran but heard nothing. Check the input device in Settings → General → Microphone.
Model <name> is already being downloaded — a second download of the same model was requested; wait for the first to finish.
Failed to load model information — Speakly could not read the models folder above.
Failed to delete model — the .bin file could not be removed. Quit Speakly, delete the file by hand, and reopen.

The model runs slowly

Pick a smaller model in Settings → Transcription — large-v3-turbo-q5_0 is the lightest option.
On a Mac with Metal, Settings → Transcription shows Apple Metal detected! Local transcription will be 3x faster on your device. with an Enable checkbox — leave it ticked.
The Audio acceleration toggle in Settings → System won't help here — it only applies to API engines; the local model always transcribes the original audio.
If the machine is simply too slow, switch to a BYOK provider in Settings → Transcription — the audio is then transcribed by the provider instead of your CPU.