Stt Uz Fastconformer Hybrid Ctc Large GGUF
GGUF conversions of the CTC branch of nvidia/sttuzfastconformerhybridlargepc for CrispASR. The upstream model is a hybrid transducer+CTC Uzbek ASR release; the shared FastConformer encoder plus the auxiliary CTC head...
Fonte do modelo
- Tipo
- Other
- Tamanho do arquivo
- 66,4–209 MB
- Modelo base
- nvidia/stt_uz_fastconformer_hybrid_large_pc
- Versão
- vRevision e077308
- Atualizado
- 26 de set. de 2026
- Downloads
- 365
- Fonte
- Hugging Face
Descrição da fonte
stt-uz-fastconformer-hybrid-ctc-large-GGUF
GGUF conversions of the CTC branch of nvidia/stt_uz_fastconformer_hybrid_large_pc for CrispASR. The upstream model is a hybrid transducer+CTC Uzbek ASR release; the shared FastConformer encoder plus the auxiliary CTC head are extracted here as a standalone CTC model (the RNNT prediction network and joint are dropped), giving a compact Uzbek ASR and forced-alignment model with punctuation + capitalisation.
| Quant | Size | Description |
|---|---|---|
| F16 | 219 MB | Full precision |
| Q8_0 | 130 MB | 8-bit |
| Q4_K | 82 MB | 4-bit K-quant (recommended) |
Architecture
17-layer NeMo FastConformer encoder + Conv1d CTC head. d_model=512, 8 heads, SentencePiece vocab, 80 log-mel features, ~115M params.
Usage
# Uzbek ASR:
crispasr --backend fastconformer-ctc -m stt-uz-fastconformer-hybrid-ctc-large-q4_k.gguf -f audio.wav
# Forced alignment (word timestamps for known text, or re-timing an .srt):
crispasr --align-only -am stt-uz-fastconformer-hybrid-ctc-large-q4_k.gguf \
-f audio.wav --text-file subtitles.srt --align-output retimed.srt
Attribution
All credit for the model goes to NVIDIA's NeMo team; this repository only repackages the CTC branch in GGUF form under the same CC-BY-4.0 license. Conversion: models/convert-stt-fastconformer-ctc-to-gguf.py in CrispASR.
Provenance and EU AI Act Art. 53 note
- Upstream model: nvidia/stt_uz_fastconformer_hybrid_large_pc — published by
nvidia. - Upstream licence:
cc-by-4.0. This repository redistributes under the same terms; it grants no rights the upstream licence does not. - What was done here: format conversion and/or quantisation only (GGUF). No training, no fine-tuning, no merging, no distillation, no change to architecture, vocabulary or capability. Only the numeric representation of the upstream weights differs.
- Training data: documented — where it is documented at all — by the upstream provider; see the upstream model card. No training data was used, added or selected by this repository.
- Provider status: under Regulation (EU) 2024/1689 the upstream authors remain the provider of this model. Converting the serialisation format does not make this repository the provider of a new general-purpose AI model, and no such claim is made. Questions about training content, copyright policy or model capability belong upstream.
Fontes
1 fonte
Fontes
1 fonteVerificado 7 de set.
Artefatos de modelo
3 artefatos
Artefatos de modelo
3 artefatosstt-uz-fastconformer-hybrid-ctc-large-f16.gguf
gguf · 209 MB · SHA-256 46f59d417e3b…0b45 · Hugging Face
Baixarstt-uz-fastconformer-hybrid-ctc-large-q4_k.gguf
gguf · 66,4 MB · SHA-256 836ab3a0933c…6fc4 · Hugging Face
Baixarstt-uz-fastconformer-hybrid-ctc-large-q8_0.gguf
gguf · 112 MB · SHA-256 93c27d3bfc61…e2de · Hugging Face
BaixarTrechos de fonte
2 trechos
Trechos de fonte
2 trechos--- license: cc-by-4.0 base_model: nvidia/stt_uz_fastconformer_hybrid_large_pc language: - uz tags: - automatic-speech-recognition - forced-alignment - gguf - crispasr - fastconformer - ctc - nemo pipeline_tag: automatic-speech-recognition --- # stt-uz-fastconformer-hybrid-ctc-large-GGUF GGUF conversions of the **CTC branch** of [nvidia/stt_uz_fastconformer_hybrid_large_pc](https://huggingface.co/nvidia/stt_uz_fastconformer_hybrid_large_pc) for [CrispASR](https://github.com/CrispStrobe/CrispASR). The upstream model is a hybrid transducer+CTC Uzbek ASR release; the shared FastConformer encoder plus the auxiliary CTC head are extracted here as a standalone CTC model (the RNNT prediction network and joint are dropped), giving a compact Uzbek ASR **and forced-alignment** model with punctuation + capitalisation. | Quant | Size | Description | |---|---|---| | F16 | 219 MB | Full precision | | Q8_0 | 130 MB | 8-bit | | Q4_K | 82 MB | 4-bit K-quant (recommended) | ## Architecture 17-layer NeMo FastConformer encoder + Conv1d CTC head. d_model=512, 8 heads, SentencePiece vocab, 80 log-mel features, ~115M params. ## Usage ```bash # Uzbek ASR: crispasr --backend fastconformer-ctc -m stt-uz-fastconformer-hybrid-ctc-large-q4_k.gguf -f audio.wav # Forced alignment (word timestamps for known text, or re-timing an .srt): crispasr --align-only -am stt-uz-fastconformer-hybrid-ctc-large-q4_k.gguf \ -f audio.wav --text-file subtitles.srt --align-output retimed.srt ``` ## Attribution All credit for the model goes to NVIDIA's NeMo team; this repository only repackages the CTC branch in GGUF form under the same CC-BY-4.0 license. Conversion: `models/convert-stt-fastconformer-ctc-to-gguf.py` in CrispASR. ## Provenance and EU AI Act Art. 53 note - **Upstream model:** [nvidia/stt_uz_fastconformer_hybrid_large_pc](https://huggingface.co/nvidia/stt_uz_fastconformer_hybrid_large_pc) — published by `nvidia`. - **Upstream licence:** `cc-by-4.0`. This repository redistributes under the same terms; it grants no rights the upstream licence does not. - **What was done here:** format conversion and/or quantisation only (GGUF). No training, no fine-tuning, no merging, no distillation, no change to architecture, vocabulary or capability. Only the numeric representation of the upstream weights differs. - **Training data:** documented — where it is documented at all — by the...
Source context: 365 downloads · 0 likes · Pipeline automatic-speech-recognition · Library nemo · Repo cstr/stt-uz-fastconformer-hybrid-ctc-large-GGUF
