Wav2Vec2-large-xlsr-hindi
theainerd/Wav2Vec2-large-xlsr-hindi
A popular open speech-to-text model, with 2.1M downloads a month. gigarouter benchmarks and hosts it as an OpenAI-compatible API.
about this model
Wav2Vec2-large-xlsr-hindi is an automatic speech recognition (ASR) model fine-tuned for Hindi. It is based on facebook/wav2vec2-large-xlsr-53 and was trained on data from the Multilingual and code-switching ASR challenges for low resource Indian languages.
Key strengths
- Specialized for Hindi speech recognition, leveraging a strong multilingual pretrained backbone.
- Designed for input sampled at 16 kHz.
Performance
Evaluated on the Hindi test split of Common Voice, the model achieves a Word Error Rate (WER) of 72.62%.
Training details
The fine-tuning script used to train the model is available as a Colab notebook.
We're benchmarking and onboarding Wav2Vec2-large-xlsr-hindi as a hosted, OpenAI-compatible API. Sign in for free credit and be ready when it lands, or tell us you want it and we'll prioritize it.