Text-to-speech Sourced statement

Kokoro 82M on Mac (Apple Silicon M1–M4)

Benchmark pending. We haven't measured this pair yet. Below is what the sources say about this hardware, quoted with links.

Unified memory shared by CPU and GPU. Runtime support varies a lot (Metal via whisper.cpp/MLX; PyTorch MPS is often partial).

EvidenceSourced statement
Commercial useYes
Licenseapache-2.0
Stated memoryNot stated
Device memory16 GB unified memory

Results

Benchmark pending. No measured numbers from us yet for Kokoro 82M on Apple Silicon. Get notified or send us your numbers (with the method).

What the sources say about Apple Silicon

  • Model authors Official kokoro README: on Mac M1/M2/M3/M4 set PYTORCH_ENABLE_MPS_FALLBACK=1 to enable GPU acceleration. source
  • Community/third-party kokoro-onnx README: 'Fast performance near real-time on macOS M1'. source

Method

We haven't run this pair. Everything on this page is quoted from the linked sources, with the source type labelled (model authors vs third parties). Vendor numbers are often measured under favourable conditions.

Ways to run Kokoro 82M

RuntimePrecisionDocumented devices
kokoro (PyTorchfp32CPU, CUDA, Apple MPS (with fallback env var)source
kokoro-onnx (onnxruntime)fp32, int8CPU (macOS M1 'near real-time' per README)source

Run it

pip install -q "kokoro>=0.9.4" soundfile
PYTORCH_ENABLE_MPS_FALLBACK=1 python your_kokoro_script.py   # see the KPipeline example in the kokoro README

Copied from the linked README. We haven't run it on this device.

New open speech models: will they run on your hardware?

One email a week: new open TTS/STT models, license gotchas, and what we measured on CPUs, Macs, Pis and consumer GPUs.