language: ja
tags:
- audio
- automatic-speech-recognition
license: apache-2.0
Kotoba-Whisper: kotoba-whisper-v1.0 for Whisper cpp
This repository contains the model weights for kotoba-tech/kotoba-whisper-v1.0 converted to GGML format. GGML is the weight format expected by C/C++ packages such as Whisper.cpp, for which we provide an example below.
Usage
Kotoba-Whisper can be run with the Whisper.cpp package with the original sequential long-form transcription algorithm.
Steps for getting started:
- Clone the Whisper.cpp repository:
git clone https://github.com/ggerganov/whisper.cpp.git
cd whisper.cpp
- Download the GGML weights for
kotoba-tech/kotoba-whisper-v1.0
:
wget https://huggingface.co/kotoba-tech/kotoba-whisper-v1.0-ggml/resolve/main/ggml-kotoba-whisper-v1.0.bin -P ./models
- Run inference using the provided sample audio:
wget https://huggingface.co/kotoba-tech/kotoba-whisper-v1.0-ggml/resolve/main/sample_ja_speech.wav
make -j && ./main -m models/ggml-kotoba-whisper-v1.0.bin -f sample_ja_speech.wav --output-file transcription --output-json
Note that it runs only with 16-bit WAV files, so make sure to convert your input before running the tool. For example, you can use ffmpeg like this:
ffmpeg -i input.mp3 -ar 16000 -ac 1 -c:a pcm_s16le output.wav
Benchmark
We measure the inference speed of different kotoba-whisper-v1.0 implementations with four different Japanese speech audio on MacBook Pro with the following spec:
- Apple M2 Pro
- 32GB
- 14-inch, 2023
- OS Sonoma Version 14.4.1 (23E224)
audio file | audio duration (min) | whisper.cpp (sec) | faster-whisper (sec) | hf pipeline (sec) |
---|---|---|---|---|
audio 1 | 50.3 | 581 | 2601 | 807 |
audio 2 | 5.6 | 41 | 73 | 61 |
audio 3 | 4.9 | 30 | 141 | 54 |
audio 4 | 5.6 | 35 | 126 | 69 |
Scripts to re-run the experiment can be found bellow:
- whisper.cpp
- faster-whisper
- hf pipeline Also, currently whisper.cpp and faster-whisper support the sequential long-form decoding, and only Huggingface pipeline supports the chunked long-form decoding, which we empirically found better than the sequnential long-form decoding.
Quantized Model
To use the quantized model, download the quantized GGML weights:
wget https://huggingface.co/kotoba-tech/kotoba-whisper-v1.0-ggml/resolve/main/ggml-kotoba-whisper-v1.0-q5_0.bin -P ./models
Run inference on the sample audio:
make -j && ./main -m models/ggml-kotoba-whisper-v1.0-q5_0.bin -f sample_ja_speech.wav --output-file transcription.quantized --output-json
Note that the benchmark results are almost identical to the raw non-quantized model weight.
Model Details
For more information about the kotoba-whisper-v1.0, refer to the original model card.