Helsinki-NLP/opus_mt - Translation Benchmark

Prerequisites

• Ubuntu 22.04 (x86_64)
• Git
• Python 3.7+
• PIP
• Picovoice Console Account

Usage

1. Clone the repository:

git clone https://github.com/Picovoice/translation-benchmark.git

2. Install the dependencies:

pip3 install -r requirements.txt

3. Set up the dataset as described in the main readme of the repository.

4. Run the benchmark:

python3 benchmark.py \
    --source-language ${SOURCE_LANGUAGE} \
    --target-language ${TARGET_LANGUAGE} \
    --engine helsinki \
    --helsinki-model-path ${HELSINI_MODEL}

Where:

• source-language is the input language.
• target-language is the output language.
• helsinki-model-path is the path to the huggingface model for the specified language pair.