voice_audio_multi/examples/xcode/Qwen3TTSDemo/README.md
Fabio c36510c21e
Some checks failed
Linux Build / Linux cpu (push) Has been cancelled
Linux Build / Linux cpu-clang (push) Has been cancelled
Linux Build / Linux vulkan (push) Has been cancelled
mac-build / macOS CPU (push) Has been cancelled
Nix Build / Nix (cpu) (push) Has been cancelled
Nix Build / Nix (vulkan) (push) Has been cancelled
windows-build / Windows CPU (push) Has been cancelled
Initial import: audio.cpp with Orange Pi documentation
2026-11-04 19:05:22 +01:00

1.4 KiB

Qwen3 TTS Xcode Demo

This is a small macOS Xcode command-line demo that links AudioCpp.xcframework and runs Qwen3 TTS voice clone or voice design from Swift through a narrow Objective-C++ bridge.

Build the framework first:

scripts/build_xcframework.sh --clean

Build the demo:

xcodebuild \
  -project examples/xcode/Qwen3TTSDemo/Qwen3TTSDemo.xcodeproj \
  -scheme Qwen3TTSDemo \
  -configuration Release \
  -derivedDataPath examples/xcode/Qwen3TTSDemo/build \
  build

Run voice clone:

examples/xcode/Qwen3TTSDemo/build/Build/Products/Release/Qwen3TTSDemo voice-clone \
  --model models/Qwen3-TTS-12Hz-1.7B-Base \
  --voice-ref resources/a.wav \
  --reference-text "This little work was finished in the year eighteen o three, and intended for immediate publication." \
  --text "The lighthouse keeper records a detailed morning report about the tide, the lantern, and the ships crossing the horizon." \
  --out /tmp/qwen3_clone.wav

Run voice design:

examples/xcode/Qwen3TTSDemo/build/Build/Products/Release/Qwen3TTSDemo voice-design \
  --model models/Qwen3-TTS-12Hz-1.7B-VoiceDesign \
  --instruct "A cheerful young woman with a bright studio voice." \
  --text "This designed voice should sound cheerful and clear while keeping every word easy to recognize." \
  --out /tmp/qwen3_design.wav

Both commands default to --backend metal --device 0 --threads 8 --language English --seed 1234 --do-sample false.