model_data_json/Amrrs_wav2vec2-large-xlsr-53-tamil.json

{
    "model_id": "Amrrs/wav2vec2-large-xlsr-53-tamil",
    "downloads": 35412,
    "tags": [
        "transformers",
        "pytorch",
        "jax",
        "wav2vec2",
        "automatic-speech-recognition",
        "audio",
        "speech",
        "xlsr-fine-tuning-week",
        "ta",
        "dataset:common_voice",
        "license:apache-2.0",
        "model-index",
        "endpoints_compatible",
        "region:us"
    ],
    "description": "--- language: ta datasets: - common_voice tags: - audio - automatic-speech-recognition - speech - xlsr-fine-tuning-week license: apache-2.0 model-index: - name: XLSR Wav2Vec2 Tamil by Amrrs results: - task: name: Speech Recognition type: automatic-speech-recognition dataset: name: Common Voice ta type: common_voice args: ta metrics: - name: Test WER type: wer value: 82.94 --- # Wav2Vec2-Large-XLSR-53-Tamil Fine-tuned facebook/wav2vec2-large-xlsr-53 in Tamil using the Common Voice When using this model, make sure that your speech input is sampled at 16kHz. ## Usage The model can be used directly (without a language model) as follows: ## Evaluation The model can be evaluated as follows on the {language} test data of Common Voice. **Test Result**: 82.94 % ## Training The Common Voice , datasets were used for training. The script used for training can be found here",
    "model_explanation_gemini": "\"Fine-tuned for Tamil automatic speech recognition using Common Voice data, achieving an 82.94% WER.\"\n\nFeatures:  \n- Language: Tamil (ta)  \n- Base Model: facebook/wav2vec2-large-xlsr-53  \n- Task: Automatic Speech Recognition (ASR)  \n- Training Data: Common Voice dataset  \n- Input Requirement: 16kHz sampled audio  \n- Metric: 82.94% Word Error Rate (WER)  \n\nComparison:",
    "release_year": null,
    "parameter_count": null,
    "is_fine_tuned": false,
    "category": "Audio",
    "model_family": "Wav2Vec",
    "api_enhanced": true
}