Downloads · 30 days
0
whisper-small-telugu-ggml
ashirbadsahu/whisper-small-telugu-ggml
whisper-small-telugu-ggml is a automatic speech recognition model from ashirbadsahu. Use it when you need speech turned into text. The card lists the license as apache-2.0.
This repository contains GGML / GGUF binary format conversions of steja/whisper-small-telugu for lightweight, high-performance CPU/GPU inference via whisper.cpp.
Access
Public
Updated Aug 18, 2026
Repo size
1.1 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin1.1 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- License
- apache-2.0
- Access
- Public
- Created
- Aug 18, 2026
- Updated
- Aug 18, 2026
- SHA
- b4e581ce
Base models
Model card
From the Hugging Face model README
whisper-small-telugu (GGML / whisper.cpp)
This repository contains GGML / GGUF binary format conversions of steja/whisper-small-telugu for lightweight, high-performance CPU/GPU inference via whisper.cpp.
Available Models
| Filename | Format | Description |
|---|---|---|
whisper_small_telugu_fp16.bin | FP16 | Full precision base GGML conversion |
whisper_small_telugu_q8_0.bin | Q8_0 | 8-bit quantized (near-zero quality loss) |
whisper_small_telugu_q5_0.bin | Q5_0 | 5-bit quantized (balanced speed/memory) |
whisper_small_telugu_q4_0.bin | Q4_0 | 4-bit quantized (minimal footprint) |
Quickstart with whisper.cpp
1. Clone and build whisper.cpp
git clone https://github.com/ggerganov/whisper.cpp.git
cd whisper.cpp
cmake -B build
cmake --build build --config Release
2. Download a model binary
# Example: Download Q5_0 quantization using huggingface-cli or curl
huggingface-cli download <your-username>/<your-repo-name> whisper_small_telugu_q5_0.bin --local-dir ./models
3. Run Inference
Make sure your audio file is a 16kHz mono WAV file:
./build/bin/whisper-cli \
-m ./models/whisper_small_telugu_q5_0.bin \
-l te \
-f /path/to/telugu_audio.wav
- Task
- Automatic Speech Recognition
- License
- apache-2.0
- Languages
- te
- Created
- Aug 18, 2026
- Updated
- Aug 18, 2026