Downloads · 30 days
7
16% of all-time downloads
CoderViking/parakeet-tdt-0.6b-v3-onnx
parakeet-tdt-0.6b-v3-onnx is a automatic speech recognition model from CoderViking. Use it when you need speech turned into text. It is set up for onnx. The card lists the license as cc-by-4.0.
int8-quantized ONNX build of nvidia/parakeet-tdt-0.6b-v3 (FastConformer encoder + Token-and-Duration Transducer decoder, 25 European languages, automatic language detection and punctuation), re-hosted for AllPrivate —…
Downloads · 30 days
7
16% of all-time downloads
All-time downloads
45
Public
Repo size
671 MB
Likes
0
Public
Click a slice to open those files.
.onnx671 MB · 100%
From the Hugging Face model README
int8-quantized ONNX build of nvidia/parakeet-tdt-0.6b-v3 (FastConformer encoder + Token-and-Duration Transducer decoder, 25 European languages, automatic language detection and punctuation), re-hosted for AllPrivate — where every model runs in the visitor's browser and nothing is uploaded.
Verified to run fully in the browser with ONNX Runtime Web (wasm execution provider): the 622 MB int8 encoder loads as a single self-contained graph (no external-data files) and produces transcriptions identical to the onnx-asr CPU reference on German and English test clips.
| file | role |
|---|---|
encoder-model.int8.onnx | FastConformer encoder, int8 QDQ, weights embedded |
decoder_joint-model.int8.onnx | prediction network + joint, int8 QDQ |
nemo128.onnx | NeMo-exact log-mel preprocessor (128 mel bins, STFT op) |
vocab.txt | 8193-entry vocabulary (token id per line, <blk> = 8192) |
config.json | model type / feature size / subsampling metadata |
fetch_and_verify.py | reproduces this repo from the pinned upstream revision |
nemo128.onnx: waveforms [B, N] f32 (16 kHz mono PCM), waveforms_lens
[B] i64 → features [B, 128, T] f32, features_lens [B] i64encoder-model.int8.onnx: audio_signal [B, 128, T] f32, length [B]
i64 → outputs [B, 1024, T′] f32, encoded_lengths [B] i64
(subsampling ×8 → one encoder frame per 80 ms)decoder_joint-model.int8.onnx: encoder_outputs [1, 1024, 1] f32,
targets [1, 1] i32, target_length [1] i32,
input_states_1/input_states_2 [2, 1, 640] f32 → outputs
[1, 1, 1, 8198] f32 (8193 token logits incl. blank, then 5 TDT duration
logits for durations 0–4), output_states_1/output_states_2argmax token if not blank (advance
LSTM states only on emission), then advance the frame index by the argmax
duration (by 1 when the duration head says 0 but the token was blank, or
after 10 emissions on one frame)model.export(), onnxruntime dynamic quantization), published
as istupakov/parakeet-tdt-0.6b-v3-onnx;
files taken unmodified from pinned revision
8f23f0c03c8761650bdb5b40aaf3e40d2c15f1cefetch_and_verify.py,
which re-downloads the upstream revision and checks the hashesonnx-asr 0.x (CPU provider) on German,
English and Spanish clips: token-identical output on the wasm execution
provider of ONNX Runtime Web 1.26CC-BY-4.0, inherited from nvidia/parakeet-tdt-0.6b-v3 (© NVIDIA). ONNX conversion by Igor Stupakov. This repo only re-packages the already-published ONNX build for browser use.