wav2vec2-large-xlsr-53-portuguese by Jonatas Grosman

SAVRN's Notes on wav2vec2-large-xlsr-53-portuguese

The page states no parameter count, so we publish no memory table: what you get is 2.5 GB of weights across bin and msgpack inside 24 files totalling 3.7 GB. It is a wav2vec2-large CTC model fine-tuned from facebook/wav2vec2-large-xlsr-53 on Portuguese — the Common Voice 6.1 train and validation splits — and it expects 16 kHz input. It draws 5.7M downloads a month.

Reported test word-error rate is 11.31, dropping to 9.01 with a language model; character error runs 3.74, or 3.21 with one — the LM plumbing is worth real points. Apache 2.0 clears commercial use. First published 2022-03-02, last updated 2022-12-14, and the XLSR-53 lineage predates the current generation of open speech models, so benchmark newer options before committing a production Portuguese pipeline.

SAVRN Research

Model Card

By Jonatas Grosman, published under apache-2.0, revision 634ac655299b.

Fine-tuned XLSR-53 large model for speech recognition in Portuguese

Fine-tuned facebook/wav2vec2-large-xlsr-53 on Portuguese using the train and validation splits of Common Voice 6.1. When using this model, make sure that your speech input is sampled at 16kHz.

This model has been fine-tuned thanks to the GPU credits generously given by the OVHcloud :)

The script used for training can be found here: https://github.com/jonatasgrosman/wav2vec2-sprint

Usage

The model can be used directly (without a language model) as follows...

Using the HuggingSound library:

from huggingsound import SpeechRecognitionModel

model = SpeechRecognitionModel("jonatasgrosman/wav2vec2-large-xlsr-53-portuguese")
audio_paths = ["/path/to/file.mp3", "/path/to/another_file.wav"]

transcriptions = model.transcribe(audio_paths)

Writing your own inference script:

Read the full model card (373 words)

Configuration

Architecture: Wav2Vec2ForCTC
Layers: 24
Hidden size: 1,024
Feed-forward size: 4,096
Attention heads: 16
Vocabulary size: 46
Model type: wav2vec2

Identity and Version

Repository: jonatasgrosman/wav2vec2-large-xlsr-53-portuguese
Publisher: Jonatas Grosman
Task: Speech recognition
Modality: Audio
Library: transformers
Parameters: Not stated by the source
Languages: pt
Revision: 634ac655299bcdc46c83bc01da9bab52d2987e4f
First published: 2022-03-02
Last updated: 2022-12-14

Files and Weights

24 files, 3.7 GB in total. The weights are 2 files totalling 2.5 GB in bin, msgpack.

Weights2 files · 2.5 GB

Configuration6 files · 8.7 KB

Tokenizer1 file · 430 B

Documentation1 file · 5.3 KB

Other13 files · 1.2 GB

Repository1 file · 736 B

Every file

File	Type	Size	SHA-256
flax_model.msgpack	Weights	1.3 GB	5601f7577c1f
pytorch_model.bin	Weights	1.3 GB	c244caf8395a
alphabet.json	Configuration	278 B	—
config.json	Configuration	1.8 KB	—
eval.py	Configuration	6.2 KB	—
language_model/attrs.json	Configuration	78 B	—
preprocessor_config.json	Configuration	262 B	—
special_tokens_map.json	Configuration	85 B	—
README.md	Documentation	5.3 KB	—
full_eval.sh	Other	1.4 KB	—
language_model/lm.binary	Other	1.2 GB	85fcd2ff940b
language_model/unigrams.txt	Other	8.1 MB	b1b261381b8d
log_mozilla-foundation_common_voice_6_0_pt_test_predictions.txt	Other	226.9 KB	—
log_mozilla-foundation_common_voice_6_0_pt_test_predictions_greedy.txt	Other	227.0 KB	—
log_mozilla-foundation_common_voice_6_0_pt_test_targets.txt	Other	228.0 KB	—
log_speech-recognition-community-v2_dev_data_pt_validation_predictions.txt	Other	121.8 KB	—
log_speech-recognition-community-v2_dev_data_pt_validation_predictions_greedy.txt	Other	122.6 KB	—
log_speech-recognition-community-v2_dev_data_pt_validation_targets.txt	Other	121.9 KB	—
mozilla-foundation_common_voice_6_0_pt_test_eval_results.txt	Other	50 B	—
mozilla-foundation_common_voice_6_0_pt_test_eval_results_greedy.txt	Other	50 B	—
speech-recognition-community-v2_dev_data_pt_validation_eval_results.txt	Other	49 B	—
speech-recognition-community-v2_dev_data_pt_validation_eval_results_greedy.txt	Other	49 B	—
.gitattributes	Repository	736 B	—
vocab.json	Tokenizer	430 B	—

License and Download

License: apache-2.0
Access: Open weights, no gate
Download size: 2.5 GB

Download from Jonatas Grosman

Released by Jonatas Grosman through its official repository on Hugging Face. Read the license.

Built From

Trained on (disclosed) common_voice
Trained on (disclosed) mozilla-foundation/common_voice_6_0

Evaluations

Each result is shown as reported, with the conditions its reporter stated. None is a SAVRN measurement. A comparison lines two results up only when their configuration, unit and setup are all stated and identical.

Benchmark	Conditions	Result	Reported by	Revision	Date
Common Voice pt	Task Automatic Speech RecognitionMetric Test CERComparison conditions not established	3.74	jonatasgrosman Publisher reported	Evaluated revision not stated	—
Common Voice pt	Task Automatic Speech RecognitionMetric Test CER (+LM)Comparison conditions not established	3.21	jonatasgrosman Publisher reported	Evaluated revision not stated	—
Common Voice pt	Task Automatic Speech RecognitionMetric Test WERComparison conditions not established	11.31	jonatasgrosman Publisher reported	Evaluated revision not stated	—
Common Voice pt	Task Automatic Speech RecognitionMetric Test WER (+LM)Comparison conditions not established	9.01	jonatasgrosman Publisher reported	Evaluated revision not stated	—
Robust Speech Event - Dev Data	Task Automatic Speech RecognitionMetric Dev CERComparison conditions not established	17.93	jonatasgrosman Publisher reported	Evaluated revision not stated	—
Robust Speech Event - Dev Data	Task Automatic Speech RecognitionMetric Dev CER (+LM)Comparison conditions not established	16.88	jonatasgrosman Publisher reported	Evaluated revision not stated	—
Robust Speech Event - Dev Data	Task Automatic Speech RecognitionMetric Dev WERComparison conditions not established	42.1	jonatasgrosman Publisher reported	Evaluated revision not stated	—
Robust Speech Event - Dev Data	Task Automatic Speech RecognitionMetric Dev WER (+LM)Comparison conditions not established	36.92	jonatasgrosman Publisher reported	Evaluated revision not stated	—

Memory Requirements

Precision	Weights in memory
As published	2.5 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About wav2vec2-large-xlsr-53-portuguese

Can I use wav2vec2-large-xlsr-53-portuguese commercially?

Yes. wav2vec2-large-xlsr-53-portuguese is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Similar Models

Model · Speech recognition

wav2vec2-large-xlsr-53-japanese

Jonatas Grosman

Fine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese using the train and validation splits of Common Voice 6.1, CSS10 and JSUT. When using this model, make sure that your speech input is sampled at 16kHz. This model has been fine-tuned thanks to the GPU credits generously given by the OVHcloud:) The script used for training can be found here: https://github.com/jonatasgrosman/wav2vec2-sprint The model can be used directly (without a language model) as follows... Using the HuggingSound library: The model can be evaluated as follows on the Japanese test data of Common Voice. In the table below I report the Word Error Rate (WER) and the Character Error Rate (CER) of the model. I ran the…

Open weights apache-2.0 transformers

View model

Model · Speech recognition

whisperkit-coreml

Argmax

WhisperKit is part of Argmax OSS, an On-device Speech AI SDK for Apple Silicon: https://github.com/argmaxinc/argmax-oss-swift Check out the WhisperKit paper and presentation from ICML 2025: https://icml.cc/virtual/2025/47854 For real-time transcription with speakers and custom vocabulary, check out Argmax Pro SDK: https://www.argmaxinc.com/blog/argmax-sdk-2

Open weights mit whisperkit

View model

Model · Speech recognition

speaker-diarization-3.1

Pyannote

Using this open-source model in production? Consider switching to pyannoteAI for better and faster options. This pipeline is the same as pyannote/speaker-diarization-3.0 except it removes the problematic use of onnxruntime. Both speaker segmentation and embedding now run in pure PyTorch. This should ease deployment and possibly speed up inference. It requires pyannote.audio version 3.1 or higher. It ingests mono audio sampled at 16kHz and outputs speaker diarization as an Annotation instance: - stereo or multi-channel audio files are automatically downmixed to mono by averaging the channels. - audio files sampled at a different rate are resampled to 16kHz automatically upon loading. 1.…

Access requested at publisher mit pyannote-audio

View model

Model · Speech recognition

speaker-diarization-community-1

Pyannote

This pipeline ingests mono audio sampled at 16kHz and outputs speaker diarization. - stereo or multi-channel audio files are automatically downmixed to mono by averaging the channels. - audio files sampled at a different rate are resampled to 16kHz automatically upon loading. The main improvements brought by Community-1 are: - improved speaker assignment and counting - simpler reconciliation with transcription timestamps with exclusive speaker diarization - easy offline use (i.e. without internet connection) - (optionally) hosted on pyannoteAI cloud 1. pip install pyannote.audio 3. Create access token at hf.co/settings/tokens. Out of the box, Community-1 is much better than…

Access requested at publisher cc-by-4.0 pyannote-audio

View model

Model · Speech recognition

faster-whisper-small

Systran

This repository contains the conversion of openai/whisper-small to the CTranslate2 model format. This model can be used in CTranslate2 or projects based on CTranslate2 such as faster-whisper. The original model was converted with the following command: Note that the model weights are saved in FP16. This type can be changed when the model is loaded using the computetype option in CTranslate2. For more information about the original model, see its model card.

Open weights mit ctranslate2

View model

Model · Speech recognition

wav2vec2-large-xlsr-53-russian

Jonatas Grosman

Fine-tuned facebook/wav2vec2-large-xlsr-53 on Russian using the train and validation splits of Common Voice 6.1 and CSS10. When using this model, make sure that your speech input is sampled at 16kHz. This model has been fine-tuned thanks to the GPU credits generously given by the OVHcloud:) The script used for training can be found here: https://github.com/jonatasgrosman/wav2vec2-sprint The model can be used directly (without a language model) as follows... Using the HuggingSound library: 1. To evaluate on mozilla-foundation/commonvoice60 with split test 2. To evaluate on speech-recognition-community-v2/devdata If you want to cite this model you can use this

Open weights apache-2.0 transformers

View model