Runs On
What it takes to serve wav2vec2-xls-r-300m-ftspeech (315M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.
SAVRN's Notes on wav2vec2-xls-r-300m-ftspeech
Transcribing the Danish parliament is what this one was tuned for. Dan Saattrup Smart fine-tuned facebook/wav2vec2-xls-r-300m on FTSpeech, 1,800 hours of transcribed speeches from the Danish parliament, into a 315M parameter CTC recognizer. The hardware question nearly answers itself: 16-bit weights are 0.6 GB and need 0.8 GB of memory, so a single MI300X at $1.85 per hour on-demand, the cheapest slot we track, is far more card than one copy needs. We would pack it beside other work on the same GPU rather than dedicate anything to it.
The license is where to slow down. It is listed as other, and the publisher says use must adhere to a license from the Danish Parliament, so read that document before any deployment. Check the publisher-reported word error rates, 13.84 on the Alvenir ASR test set and 17.91 on Danish Common Voice 8.0, against your own audio before committing.
SAVRN Research, 2026-09-18
Model Card
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the FTSpeech dataset, being a dataset of 1,800 hours of transcribed speeches from the Danish parliament. The model achieves the following WER scores (lower is better): The use of this model needs to adhere to this license from the Danish Parliament.
Excerpt from the card by Dan Saattrup Smart, licensed other.
Configuration
- Architecture
- Wav2Vec2ForCTC
- Layers
- 24
- Hidden size
- 1,024
- Feed-forward size
- 4,096
- Attention heads
- 16
- Vocabulary size
- 35
- Stored precision
- float32
- Model type
- wav2vec2
Identity and Version
- Repository
- saattrupdan/wav2vec2-xls-r-300m-ftspeech
- Publisher
- Dan Saattrup Smart
- Task
- Speech recognition
- Modality
- Audio
- Library
- transformers
- Parameters
- 315M parameters
- Languages
- da
- Revision
- 7a60985d596e7a9e26b01a0707fe59daac6ffe84
- First published
- 2022-03-04
- Last updated
- 2023-09-11
Files and Weights
15 files, 3.8 GB in total. The weights are 4 files totalling 3.8 GB in bin, safetensors.
Weights4 files · 3.8 GB
Configuration5 files · 2.7 KB
Tokenizer2 files · 666 B
Documentation1 file · 1.5 KB
Other1 file · 7.9 MB
Repository2 files · 1.2 KB
Every file
License and Download
- License
- other
- Access
- Open weights, no gate
- Download size
- 3.8 GB
Built From
- Derived from
facebook/wav2vec2-xls-r-300m
- Trained on (disclosed)
ftspeech
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.
Memory Requirements
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Compare wav2vec2-xls-r-300m-ftspeech
Questions About wav2vec2-xls-r-300m-ftspeech
How much GPU memory does wav2vec2-xls-r-300m-ftspeech need?
About 0.8 GB at 16-bit and 0.2 GB at 4-bit: the weights (315M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run wav2vec2-xls-r-300m-ftspeech on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
What license is wav2vec2-xls-r-300m-ftspeech released under?
other, as its publisher declares it. Read the license text before commercial use.
Similar Models
You can test this model online with the Space for Romanian Speech Recognition The model ranked TOP-1 on Romanian Speech Recognition during HuggingFace's Robust Speech Challenge: This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the Common Voice 8.0 - Romanian subset dataset, with extra training data from Romanian Speech Synthesis dataset. Without the 5-gram Language Model optimization, it achieves the following results on the evaluation set (Common Voice 8.0, Romanian subset, test split): The architecture is based on facebook/wav2vec2-xls-r-300m with a speech recognition CTC head and an added 5-gram language model (using pyctcdecode and kenlm) trained on the Romanian…
Open weights
apache-2.0
315M parameters
transformers
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the private datasets in 2 stages - firstly was fine-tuned on a small dataset with good samples Then the obtained model was fine-tuned on a large dataset with the small good dataset, with various samples from different sources, and with an unlabeled dataset that was weakly labeled using a previously trained model. (weakly labeled data wasn't used in validation set) on small dataset on large dataset on small dataset on large dataset The following hyperparameters were used during training: - learningrate: 0.0003 - trainbatchsize: 8 - evalbatchsize: 8 - distributedtype: multi-GPU - numdevices: 2 - gradientaccumulationsteps: 4…
Open weights
315M parameters
transformers
This Python package provides an efficient way to perform forced alignment between text and audio using Hugging Face's pretrained models. it also features an improved implementation to use much less memory than TorchAudio forced alignment API. The model checkpoint uploaded here is a conversion from torchaudio to HF Transformers for the MMS-300M checkpoint trained on forced alignment dataset
Open weights
cc-by-nc-4.0
315M parameters
transformers
ATTENTION! Metrics (float16) using evaluate library with batchsize=1
Open weights
apache-2.0
315M parameters
transformers
Finetuned version of KBs VoxRex large model using Swedish radio broadcasts, NST and Common Voice data. Evalutation without a language model gives the following: WER for NST + Common Voice test set (2% of total sentences) is 2.5%. WER for Common Voice test set is 8.49% directly and 7.37% with a 4-gram language model. When using this model, make sure that your speech input is sampled at 16kHz. This model has been fine-tuned for 120000 updates on NST + CommonVoice and then for an additional 20000 updates on CommonVoice only. The additional fine-tuning on CommonVoice hurts performance on the NST+CommonVoice test set somewhat and, unsurprisingly, improves it on the CommonVoice test set. It seems…
Open weights
cc0-1.0
315M parameters
transformers
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the commonvoice 8.0 dataset as well as other datasets listed below. It achieves the following results on the evaluation set: The eval.py script results using a LM are: Fine-tuned facebook/wav2vec2-large-xlsr-53 on Czech using the Common Voice dataset. When using this model, make sure that your speech input is sampled at 16kHz. The model can be used directly (without a language model) as follows: The model can be evaluated using the attached eval.py script: The Common Voice 8.0 train and validation datasets were used for training, as well as the following datasets: - Šmídl, Luboš and Pražák, Aleš, 2013, OVM – Otázky…
Open weights
apache-2.0
315M parameters
transformers
