README.md

---
license: apache-2.0
tags:
  - coreml
  - tts
  - text-to-speech
  - apple
  - qwen3
language:
  - en
  - zh
library_name: coremltools
---

# Qwen3-TTS CoreML

CoreML conversion of [Qwen/Qwen3-TTS](https://huggingface.co/Qwen/Qwen3-TTS) (0.6B) for on-device inference on Apple platforms.

Supports English and Chinese text-to-speech synthesis.

## Models

| Model | Description | Size |
|-------|-------------|------|
| `qwen3_tts_lm_prefill_v9` | LM KV-cache prefill (text + speaker conditioning) | ~2.8 GB |
| `qwen3_tts_lm_decode_v10` | Autoregressive LM decode (CB0 codec token generation) | ~1.8 GB |
| `qwen3_tts_cp_prefill` | Code predictor prefill (CB1-15 conditioning) | ~432 MB |
| `qwen3_tts_cp_decode` | Code predictor decode (CB1-15 generation) | ~420 MB |
| `qwen3_tts_decoder_10s` | Audio decoder (16-codebook codes → 24kHz waveform) | ~436 MB |
| `speaker_embedding_official.npy` | Default speaker embedding (1024-dim) | 4 KB |

**Total: ~5.9 GB**

## Pipeline

```
Text tokens + Speaker embedding
        ↓
  LM Prefill (KV cache initialization)
        ↓
  LM Decode (CB0 codec tokens, temperature=0.9, top_k=50)
        ↓
  Code Predictor Prefill + Decode (CB1-15 per frame)
        ↓
  Audio Decoder (16 codebooks → 24kHz waveform)
        ↓
  Silence trimming → Final audio
```

## Key Parameters

- **Sample rate:** 24,000 Hz
- **Codebooks:** 16 (CB0 from LM, CB1-15 from code predictor)
- **Max codec tokens:** 125 frames (~10s audio)
- **Sampling:** temperature=0.9, top_k=50 (both CB0 and CB1-15)
- **EOS token ID:** 2150 (in codec logit space)

## Usage

```swift
import FluidAudioTTS

let manager = Qwen3TtsManager()
try await manager.loadFromDirectory(modelDir)

let wav = try await manager.synthesize(
    text: "Hello world",
    tokenIds: [9707, 1879, ...],  // Pre-tokenized with Qwen3 processor
    useSpeaker: true
)
```

See [FluidAudio](https://github.com/FluidInference/FluidAudio) for the full Swift framework.

## Conversion

Converted using [coremltools](https://github.com/apple/coremltools) from the original PyTorch weights. Conversion scripts are in the [mobius](https://github.com/FluidInference/mobius) repository.

## License

**Apache-2.0**, inherited from [Qwen/Qwen3-TTS](https://huggingface.co/Qwen/Qwen3-TTS).

## References

- [Qwen/Qwen3-TTS](https://huggingface.co/Qwen/Qwen3-TTS)
- [FluidAudio](https://github.com/FluidInference/FluidAudio)
- [mobius](https://github.com/FluidInference/mobius)