Roest Wav2vec2 315m V2

🚀 Røst-wav2vec2-315m-v2

This is a Danish state-of-the-art speech recognition model, trained as part of the CoRal project by Alvenir. It's designed for automatic speech recognition (ASR), leveraging the rich CoRal-v2 dataset which includes diverse Danish conversational and read - aloud data across different age groups, genders, and dialects. Try it out in our interactive demo!

🚀 Quick Start

Installation

First, install the necessary libraries:

$ pip install transformers kenlm pyctcdecode

Usage

You can use the model with the transformers Python package as follows:

>>> from transformers import pipeline
>>> audio = get_audio()  # 16kHz raw audio array
>>> transcriber = pipeline(model="CoRal-project/roest-wav2vec2-315m-v2")
>>> transcriber(audio)
{'text': 'your transcription'}

✨ Features

• Rich Training Data: Trained on the CoRal-v2 dataset, which contains a wide range of Danish speech data, including various dialects, age groups, and genders.
• ASR Focus: Specifically designed for automatic speech recognition, aiming to accurately transcribe Danish speech.
• Interactive Demo: Allows users to easily test the model's performance through our interactive demo.

💻 Usage Examples

Basic Usage

The following code shows how to use the model for basic transcription:

>>> from transformers import pipeline
>>> audio = get_audio()  # 16kHz raw audio array
>>> transcriber = pipeline(model="CoRal-project/roest-wav2vec2-315m-v2")
>>> transcriber(audio)
{'text': 'your transcription'}

Transcription Examples with Different Dialects

Explore the following audio samples along with their transcriptions and accuracy metrics. Each example showcases the model's performance with different Danish dialects.

📚 Documentation

Model Details

Wav2Vec2 is a state-of-the-art model architecture for speech recognition, leveraging self-supervised learning from raw audio data. The pre-trained Wav2Vec2-XLS-R-300M has been fine-tuned for automatic speech recognition with the CoRal-v2 dataset dataset to enhance its performance in recognizing Danish speech with consideration to different dialects. The model was trained for 30K steps using the training setup in the CoRaL repository by running:

python src/scripts/finetune_asr_model.py \
  model=wav2vec2-small \
  max_steps=30000 \
  datasets.coral_conversation_internal.id=CoRal-project/coral-v2 \
  datasets.coral_readaloud_internal.id=CoRal-project/coral-v2

The model is evaluated using a Language Model (LM) as post-processing. The utilized LM is the one trained and used by CoRal-project/roest-wav2vec2-315m-v1.

The model was trained on the CoRal-v2 dataset, including both the conversational and read-aloud subset. This dataset consists of Danish speech across a variety of dialects, age groups and gender distinctions. Note that the dataset, and thus also this model, is licensed under a custom license, adapted from OpenRAIL-M, which allows commercial use with few restrictions (speech synthesis and biometric identification) - see license.

Evaluation

The model was evaluated using the following metrics:

• Character Error Rate (CER): The percentage of characters incorrectly transcribed.
• Word Error Rate (WER): The percentage of words incorrectly transcribed.

Conversational CoRal Performance

The model was firstly evaluated on a tentative version of the coral-v2 conversation dataset.

The results are tentative as the test set only includes 5 unique speakers, of which 4 are women. The test set includes 2 speakers with 'Fynsk' dialect, 1 with 'Sønderjysk', 1 with 'Non-native' and 1 'Nordjysk'.

Note that the high generalization error on conversation data for models trained on read-aloud data is still being analyzed.

Read-aloud CoRal Performance

OBS! Benchmark for hviske-v2 has been re-evaluated and the confidence interval is larger than reported in the model card.

Performance on Other Datasets

The model was also tested against other datasets to evaluate generalizability:

OBS! The vocab used for training incudes numerals (0,1,2,..,9), which are translated to text in a post-processing step. If the model misses spaces the numbers are interpreted as one, which especially affects the NST score as this dataset contains many numerals.

Note on comparing Whisper and Wav2Vec2 models

The Whisper models detailed in this model card exhibit significantly lower Character Error Rates (CER) and Word Error Rates (WER) compared to the Wav2Vec2 models. Whisper utilizes a transformer-based architecture with additional layers that enhance contextual understanding. In contrast, Wav2Vec2 models employ shorter context windows that focus on sound prediction. The Røst-Wav2Vec2 models incorporate a straightforward language model during post-processing, which addresses errors based on statistical language patterns. Introducing a more complex, contextual post-processing language model might enable a better comparison between these model types, which the CoRal project plans to explore in future releases.

The Røst-Whisper model excels in read-aloud data, leveraging its embedded contextual framework to achieve more robust recognition within this context. However, Wav2Vec2 models appear to generalize more effectively across various speech recognition tasks, whereas Whisper models incur higher error rates in conversational data. It’s important to note that the CoRal-v2 conversation dataset, being tentative and featuring limited speaker diversity, might influence these results.

🔧 Technical Details

Training

• Base Model: Wav2Vec2-XLS-R-300M
• Training Dataset: CoRal-v2, including both conversational and read-aloud subsets.
• Training Steps: 30K steps
• Training Command:

python src/scripts/finetune_asr_model.py \
  model=wav2vec2-small \
  max_steps=30000 \
  datasets.coral_conversation_internal.id=CoRal-project/coral-v2 \
  datasets.coral_readaloud_internal.id=CoRal-project/coral-v2

Evaluation

• Metrics: Character Error Rate (CER) and Word Error Rate (WER)
• Language Model: A Language Model (LM) is used for post-processing. The utilized LM is the one trained and used by CoRal-project/roest-wav2vec2-315m-v1.

📄 License

The model is licensed under a custom license, adapted from OpenRAIL-M, which allows commercial use with few restrictions (speech synthesis and biometric identification). See license for more details.

Training curves

Creators and Funders

This model has been trained and the model card written by Marie Juhl Jørgensen and Søren Vejlgaard Holm at Alvenir.

The CoRal project is funded by the Danish Innovation Fund and consists of the following partners:

• Alexandra Institute
• University of Copenhagen
• Agency for Digital Government
• Alvenir
• Corti

We would like specifically to thank Dan Saattrup Nielsen, Alexandra Institute for (among other things) the repository work and Simon Leminen Madsen, Alexandra Institute for modelling work.

Citation

@misc{roest-wav2vec2-315m-v2,
  author    = {Marie Juhl Jørgensen, Søren Vejlgaard Holm, Martin Carsten Nielsen, Dan Saattrup Nielsen, Sif Bernstorff Lehmann, Simon Leminen Madsen and Torben Blach},
  title     = {Røst-wav2vec-315m-v2: A Danish state-of-the-art speech recognition model trained on varied demographics and dialects},
  year      = {2025},
  url       = {https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2},
}