parakeet_de_med is Johann Hartmann's PEFT decoder+joint fine-tune of nvidia/parakeet-tdt-0.6b-v3 specialised for German medical documentation (Arztbriefe). On the German medical test set it scores 3.28% WER vs the base model's 11.73% — a 72% relative reduction.
The fine-tune freezes the encoder and trains only the TDT decoder + joint head (18.1M out of 627M parameters, 2.89%). This means:
The architecture is identical to parakeet-tdt-0.6b-v3 (24-layer FastConformer encoder, 2-layer LSTM predictor, 8198-class TDT joint head)
The same GGUF converter, runtime, and CLI work as-is
The frozen encoder still uses the base model's auto-language detection — for clean German speech this works well, for accented or noisy audio you may want to fall back to a different runtime (see comparison table)
Files
File
Size
Notes
parakeet_de_med.gguf
1.26 GB
F16, full precision
parakeet_de_med-q8_0.gguf
711 MB
Q8_0, near-lossless
parakeet_de_med-q5_0.gguf
516 MB
Q5_0
parakeet_de_med-q4_k.gguf
467 MB
Q4_K — recommended default
All quantisations produce the same text on the German verification clip:
The runtime is the same parakeet-main binary used for the base parakeet-tdt-0.6b-v3. All the usual flags work: -vad-model for Silero VAD slicing, -ck N for fixed chunking, -ml N for max chars per line, -osrt/-ovtt/-ot for subtitle output, -v for per-token timestamps via the TDT duration head.
Word-level timestamps
Like the base parakeet model, this fine-tune emits TDT durations as part of decoding, so word-level timestamps come for free at one encoder frame = 80 ms granularity. No separate forced alignment model needed:
Fine-tune:johannhartmann/parakeet_de_med (CC-BY-4.0). Johann Hartmann. Trained on 976 German medical documentation samples for 5 epochs with PEFT decoder+joint strategy.
CC-BY-4.0, inherited from both the base model and the fine-tune. Use of these GGUF files must comply with the CC-BY-4.0 license including attribution to NVIDIA NeMo team and Johann Hartmann.
