Downloads · 30 days
17
7% of all-time downloads
voiceflow-corrector-llama-3.2-3b-gguf
VikramKavur/voiceflow-corrector-llama-3.2-3b-gguf
voiceflow-corrector-llama-3.2-3b-gguf is a machine learning model from VikramKavur. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as llama3.2.
A LoRA-SFT finetune of Llama-3.2-3B-Instruct (merged, quantized to Q4KM) that corrects speech-to-text transcripts: homophones, word boundaries, punctuation, casing, and spoken numbers. Built for the VoiceFlow desktop…
All-time downloads
251
Public
Repo size
2 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf2 GB · 100%
At a glance
- License
- llama3.2
- Access
- Public
- Created
- Jun 20, 2026
- Updated
- Jun 20, 2026
- SHA
- 598b8820
Base models
Model card
From the Hugging Face model README
VoiceFlow Corrector — Llama 3.2 3B (finetuned, GGUF)
A LoRA-SFT finetune of Llama-3.2-3B-Instruct (merged, quantized to Q4_K_M) that corrects speech-to-text transcripts: homophones, word boundaries, punctuation, casing, and spoken numbers. Built for the VoiceFlow desktop app.
- Base model: meta-llama/Llama-3.2-3B-Instruct
- Method: LoRA SFT on ~3.3k ASR-correction pairs, merged + Q4_K_M
- Prompt: bare
transcript → corrected(no verbose system prompt) - Use: load with llama.cpp / llama-cpp-python; greedy decoding (temperature 0)
License
Governed by the Llama 3.2 Community License.
- License
- llama3.2
- Created
- Jun 20, 2026
- Updated
- Jun 20, 2026