The jonatasgrosman/wav2vec2-large-xlsr-53-dutch model performs automatic speech recognition for Dutch. It converts spoken audio into text and forms part of the wav2vec 2.0 family of models that have been fine-tuned on Dutch data from the Common Voice corpus.
The model is implemented in the Transformers library and supports direct loading through AutoProcessor and AutoModelForCTC classes or via a high-level pipeline for automatic-speech-recognition tasks. It carries an Apache 2.0 license and is hosted on the Hugging Face platform, where it can be used with PyTorch or JAX backends. The training drew from the mozilla-foundation/common_voice_6_0 dataset and participated in the xlsr-fine-tuning-week and related speech recognition evaluation events.
Developers and researchers integrate the model into applications that require transcription of Dutch audio. Access occurs through the Hugging Face ecosystem, including inference providers, notebooks, and local installations of the Transformers library. The page provides code examples for both pipeline and direct model loading to enable immediate use in speech-to-text workflows.
Wav2vec2 Large Xlsr 53 Dutch is a Voice, TTS & speech product. It focuses on transcribing spoken Dutch audio into text automatically for various applications. It is built as an open-source project for speech and NLP developers. Wav2vec2 Large Xlsr 53 Dutch is open source under the Open Source license. It runs on the web, the command line, and API, and it can be self-hosted.
jonatasgrosman builds and maintains Wav2vec2 Large Xlsr 53 Dutch, and the product first shipped in 2022. Key capabilities include speech recognition, dutch language support, and automatic transcription.
Latest indexed changes and source events
jonatasgrosman/wav2vec2-large-xlsr-53-dutch verified by the PulseGate indexer
Other apps tracked under the same category.
