Downloads · 30 days
73
4% of all-time downloads
parakeet-tdt-0.6b-v3
sonic-speech/parakeet-tdt-0.6b-v3
parakeet-tdt-0.6b-v3 is a automatic speech recognition model from sonic-speech. Use it when you need speech turned into text. It is set up for mlx. The card lists the license as cc-by-4.0.
NVIDIA Parakeet TDT v3 automatic speech recognition model in MLX BF16 SafeTensors format for Apple Silicon. This is the reference BF16 checkpoint — see quantized variants for reduced memory and faster inference:
All-time downloads
1.9K
Public
Parameters
627M
2.5 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors2.5 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- mlx
- License
- cc-by-4.0
- Access
- Public
- Created
- Jan 8, 2026
- Updated
- Mar 8, 2026
- SHA
- 55c316ec
Base models
Model card
From the Hugging Face model README
Parakeet TDT 0.6B v3 (MLX, BF16)
NVIDIA Parakeet TDT v3 automatic speech recognition model in MLX BF16 SafeTensors format for Apple Silicon. This is the reference BF16 checkpoint — see quantized variants for reduced memory and faster inference:
sonic-speech/parakeet-tdt-0.6b-v3-int8— Encoder INT8 (recommended)sonic-speech/parakeet-tdt-0.6b-v3-int4— Encoder INT4 (lite, for 8GB Macs)
Benchmark Results (M3 Max, 64GB)
Variant Comparison
| Variant | Size | WER (LibriSpeech) | WER (TED-LIUM) | RTFx | Peak Memory |
|---|---|---|---|---|---|
| BF16 | 1,254 MB | 0.82% | 15.1% | 73x | 3,002 MB |
| INT8 | 755 MB | 0.82% | 15.1% | 95x | 1,268 MB |
| INT4 | 489 MB | 0.82% | 15.5% | 98x | 1,003 MB |
- LibriSpeech test-clean: 50 samples, studio-quality read speech
- TED-LIUM: 8 TED talks (60s segments), real-world acoustics
RTFx vs Audio Duration
| 5s | 15s | 30s | 60s | 120s |
|---|---|---|---|---|
| 59x | 91x | 98x | 103x | 99x |
Quantization Strategy
Encoder-only mixed-precision: the Conformer encoder (~85% of parameters) is quantized while the decoder and joint network remain BF16. This preserves decoder precision for rare words and punctuation.
| Variant | Size Reduction | Speed Gain | Memory Reduction | WER Impact |
|---|---|---|---|---|
| INT8 | -40% | +30% | -58% | None |
| INT4 | -61% | +34% | -67% | +0.4pp on real speech |
Usage
Install:
Model Details
- Architecture: Conformer encoder + TDT (Token-and-Duration Transducer) decoder
- Parameters: 627M
- Languages: 25 (SentencePiece tokenizer)
- Sample rate: 16 kHz
- Precision: BF16 (optimized for Apple Silicon)
Origin
Weights from mlx-community/parakeet-tdt-0.6b-v3, converted from NVIDIA's official nvidia/parakeet-tdt-0.6b-v3.
Part of the Sonic Speech model collection for the Sonic local-first voice AI project.
- Task
- Automatic Speech Recognition
- Library
- mlx
- License
- cc-by-4.0
- Languages
- en
- Created
- Jan 8, 2026
- Updated
- Mar 8, 2026