MLX Speech Models

aufklarer 's Collections

MLX Speech Models

CoreML Speech Models

Speech Android Models

updated 1 day ago

Speech AI models for Apple Silicon via MLX. ASR, TTS, VAD, diarization, speaker embedding.

Upvote

aufklarer/WeSpeaker-ResNet34-LM-MLX

Audio Classification • Updated Apr 12 • 366k • 2
aufklarer/Qwen3-ASR-0.6B-MLX-4bit

0.3B • Updated Apr 12 • 57.8k • 2
aufklarer/Qwen3-ForcedAligner-0.6B-4bit

Audio Classification • Updated Apr 12 • 50.6k • 1
aufklarer/Pyannote-Segmentation-MLX

Voice Activity Detection • Updated Apr 12 • 6.53k
aufklarer/Qwen3-ASR-1.7B-MLX-8bit

0.8B • Updated Apr 12 • 6.26k • 2
aufklarer/Silero-VAD-v5-MLX

Voice Activity Detection • Updated Apr 12 • 6.14k • 3
aufklarer/Qwen3-ASR-0.6B-MLX-8bit

0.4B • Updated Apr 12 • 1.22k
aufklarer/PersonaPlex-7B-MLX-4bit

Audio-to-Audio • Updated Apr 12 • 669 • 32
aufklarer/Qwen3-ASR-1.7B-MLX-4bit

Automatic Speech Recognition • Updated Apr 12 • 822 • 1
aufklarer/PersonaPlex-7B-MLX-8bit

Audio-to-Audio • Updated Apr 12 • 412 • 6
aufklarer/Qwen3-TTS-12Hz-0.6B-Base-MLX-4bit

0.4B • Updated Apr 12 • 409
aufklarer/CosyVoice3-0.5B-MLX-4bit

Text-to-Speech • Updated 1 day ago • 340
aufklarer/Qwen3-ForcedAligner-0.6B-8bit

Audio Classification • Updated Apr 12 • 275
aufklarer/Qwen3-ForcedAligner-0.6B-bf16

Audio Classification • Updated Apr 12 • 228
aufklarer/Qwen3-TTS-12Hz-1.7B-Base-MLX-8bit

0.8B • Updated Apr 12 • 159
aufklarer/Qwen3-TTS-12Hz-0.6B-CustomVoice-MLX-4bit

0.4B • Updated Apr 12 • 145
aufklarer/Qwen3-TTS-12Hz-1.7B-Base-MLX-4bit

0.6B • Updated Apr 12 • 114
aufklarer/Qwen3-TTS-12Hz-0.6B-Base-MLX-8bit

0.5B • Updated Apr 12 • 53
aufklarer/Omnilingual-ASR-CTC-7B-MLX-8bit

Automatic Speech Recognition • Updated Apr 12 • 31
aufklarer/Omnilingual-ASR-CTC-300M-MLX-4bit

Automatic Speech Recognition • Updated Apr 12 • 39
aufklarer/Omnilingual-ASR-CTC-3B-MLX-8bit

Automatic Speech Recognition • Updated Apr 12 • 13
aufklarer/Omnilingual-ASR-CTC-1B-MLX-8bit

Automatic Speech Recognition • Updated Apr 12 • 11
aufklarer/OpenUnmix-HQ-MLX

Audio-to-Audio • Updated Mar 25 • 20
aufklarer/Qwen3.5-0.8B-Chat-MLX

Text Generation • Updated Apr 12
aufklarer/Omnilingual-ASR-CTC-3B-MLX-4bit

Automatic Speech Recognition • Updated Apr 12 • 11
aufklarer/Omnilingual-ASR-CTC-1B-MLX-4bit

Automatic Speech Recognition • Updated Apr 12 • 16
aufklarer/OpenUnmix-L-MLX

Audio-to-Audio • Updated Mar 25 • 2
aufklarer/Omnilingual-ASR-CTC-300M-MLX-8bit

Automatic Speech Recognition • Updated Apr 12 • 13
aufklarer/Omnilingual-ASR-CTC-7B-MLX-4bit

Automatic Speech Recognition • Updated Apr 12 • 157
aufklarer/VibeVoice-Realtime-0.5B-MLX-INT4

Text-to-Speech • Updated 19 days ago • 78
aufklarer/VibeVoice-Realtime-0.5B-MLX-INT8

Text-to-Speech • Updated 19 days ago • 93
aufklarer/VibeVoice-1.5B-MLX-INT4

Text-to-Speech • Updated 19 days ago • 188
aufklarer/MADLAD400-3B-MT-MLX

Translation • Updated 18 days ago

Note Many-to-many translation across 400+ languages (T5 v1.1, INT4/INT8).
aufklarer/CosyVoice3-0.5B-MLX-8bit

Text-to-Speech • Updated 1 day ago • 24

Note 8-bit LLM variant — bundled S3-Tokenizer-v3 for zero-shot voice cloning

Upvote