Qwen3 ForcedAligner 0.6B CoreML INT4
CoreML conversion of Qwen/Qwen3-ForcedAligner-0.6B with INT4 palettization for Apple Neural Engine.
Modellquelle
- Art
- Other
- Dateigröße
- 151–297 MB
- Basismodell
- Qwen/Qwen3-ForcedAligner-0.6B
- Version
- vRevision bb755f7
- Aktualisiert
- 5. Nov. 2026
- Downloads
- 46
- Quelle
- Hugging Face
Quellenbeschreibung
Qwen3-ForcedAligner-0.6B — CoreML INT4
CoreML conversion of Qwen/Qwen3-ForcedAligner-0.6B with INT4 palettization for Apple Neural Engine.
Predicts word-level timestamps in a single forward pass.
Models
| Model | Description | Quantization |
|---|---|---|
encoder.mlmodelc | Audio encoder (24 layers) | INT4 palettized |
decoder.mlmodelc | Text decoder + classify head (28 layers) | INT4 palettized |
Usage
let aligner = try await CoreMLForcedAligner.fromPretrained(
modelId: "aufklarer/Qwen3-ForcedAligner-0.6B-CoreML-INT4"
)
let aligned = aligner.align(audio: samples, text: "Hello world", sampleRate: 24000)
Variants
| Variant | Backend | Size | Model ID |
|---|---|---|---|
| CoreML INT4 | Neural Engine | ~630 MB | aufklarer/Qwen3-ForcedAligner-0.6B-CoreML-INT4 |
| CoreML INT8 | Neural Engine | ~1.0 GB | aufklarer/Qwen3-ForcedAligner-0.6B-CoreML-INT8 |
| MLX 4-bit | GPU | ~979 MB | aufklarer/Qwen3-ForcedAligner-0.6B-4bit |
| MLX 8-bit | GPU | ~1.4 GB | aufklarer/Qwen3-ForcedAligner-0.6B-8bit |
Links
- Swift library: soniqo/speech-swift
- Base model: Qwen/Qwen3-ForcedAligner-0.6B
- Guide: soniqo.audio/guides/align
- Docs: soniqo.audio
- GitHub: soniqo/speech-swift
Quellen
1 Quelle
Quellen
1 QuelleVerifiziert 5. Nov.
Download files
Repository files
File inventory only; completeness is unverified.
- audio_encoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin
- audio_encoder.mlpackage/Manifest.json
- config.json
- embedding.mlpackage/Data/com.apple.CoreML/weights/weight.bin
- embedding.mlpackage/Manifest.json
- merges.txt
- text_decoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin
- text_decoder.mlpackage/Manifest.json
- tokenizer_config.json
- vocab.json
Modellartefakte
3 Artefakte
Modellartefakte
3 Artefakteaudio_encoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin
bin · 151 MB · SHA-256 922194bb1248…2cde · Hugging Face
Download fileembedding.mlpackage/Data/com.apple.CoreML/weights/weight.bin
bin · 297 MB · SHA-256 3c6da8aa0678…d8cd · Hugging Face
Download filetext_decoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin
bin · 213 MB · SHA-256 09a515633804…88e7 · Hugging Face
Download fileQuellenauszüge
3 Auszüge
Quellenauszüge
3 Auszüge--- license: apache-2.0 language: - en - zh - ja - ko - de - fr - es - it - ru tags: - coreml - forced-alignment - speech - qwen3 - timestamps - neural-engine base_model: Qwen/Qwen3-ForcedAligner-0.6B pipeline_tag: audio-classification --- # Qwen3-ForcedAligner-0.6B — CoreML INT4 CoreML conversion of [Qwen/Qwen3-ForcedAligner-0.6B](https://huggingface.co/Qwen/Qwen3-ForcedAligner-0.6B) with INT4 palettization for Apple Neural Engine. Predicts word-level timestamps in a single forward pass. ## Models | Model | Description | Quantization | |-------|-------------|-------------| | `encoder.mlmodelc` | Audio encoder (24 layers) | INT4 palettized | | `decoder.mlmodelc` | Text decoder + classify head (28 layers) | INT4 palettized | ## Usage ```swift let aligner = try await CoreMLForcedAligner.fromPretrained( modelId: "aufklarer/Qwen3-ForcedAligner-0.6B-CoreML-INT4" ) let aligned = aligner.align(audio: samples, text: "Hello world", sampleRate: 24000) ``` ## Variants | Variant | Backend | Size | Model ID | |---------|---------|------|----------| | **CoreML INT4** | **Neural Engine** | **~630 MB** | **aufklarer/Qwen3-ForcedAligner-0.6B-CoreML-INT4** | | CoreML INT8 | Neural Engine | ~1.0 GB | [aufklarer/Qwen3-ForcedAligner-0.6B-CoreML-INT8](https://huggingface.co/aufklarer/Qwen3-ForcedAligner-0.6B-CoreML-INT8) | | MLX 4-bit | GPU | ~979 MB | [aufklarer/Qwen3-ForcedAligner-0.6B-4bit](https://huggingface.co/aufklarer/Qwen3-ForcedAligner-0.6B-4bit) | | MLX 8-bit | GPU | ~1.4 GB | [aufklarer/Qwen3-ForcedAligner-0.6B-8bit](https://huggingface.co/aufklarer/Qwen3-ForcedAligner-0.6B-8bit) | ## Links - **Swift library**: [soniqo/speech-swift](https://github.com/soniqo/speech-swift) - **Base model**: [Qwen/Qwen3-ForcedAligner-0.6B](https://huggingface.co/Qwen/Qwen3-ForcedAligner-0.6B) --- --- - **Guide**: [soniqo.audio/guides/align](https://soniqo.audio/guides/align) - **Docs**: [soniqo.audio](https://soniqo.audio) - **GitHub**: [soniqo/speech-swift](https://github.com/soniqo/speech-swift)
Source context: 46 downloads · 0 likes · Pipeline audio-classification · Repo aufklarer/Qwen3-ForcedAligner-0.6B-CoreML-INT4
Source context: 20 downloads · 0 likes · Pipeline audio-classification · Repo aufklarer/Qwen3-ForcedAligner-0.6B-CoreML-INT4
