ArttuPakarinen/sesame-csm-FIN-parlament-full-finetune
2B • Updated • 1 • 1
You need to agree to share your contact information to access this dataset
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
Terms of Access: The researcher has requested permission to use the Nord-Parl-TTS dataset. In exchange for such permission, the researcher hereby agrees to the following terms and conditions:
- The researcher shall use the Nord-Parl-TTS dataset under the CC-BY-NC-ND license. 2. The authors make no representations or warranties regarding the datasets,
including but not limited to warranties of non-infringement or fitness for
a particular purpose. - The researcher accepts full responsibility for their use of the datasets and
shall defend and indemnify the authors of Nord-Parl-TTS, including their employees, trustees, officers, and agents,
against any and all claims arising from the researcher's use of the datasets,
including but not limited to the researcher's use of any copies of copyrighted
content that they may create from the datasets. - The researcher may provide research associates and colleagues with access
to the datasets, provided that they first agree to be bound by these terms
and conditions. - The authors reserve the right to terminate the researcher's access to the
datasets at any time. - If the researcher is employed by a for-profit, commercial entity, the
researcher's employer shall also be bound by these terms and conditions,
and the researcher hereby represents that they are fully authorized to enter
into this agreement on behalf of such employer.
Log in or Sign Up to review the conditions and access this dataset content.
Nord-Parl-TTS: Finnish and Swedish TTS Dataset from Parliament Speech
News
- 2026.01.17: 🎉 Our paper "Nord-Parl-TTS: Finnish and Swedish TTS Dataset from Parliament Speech" has been accepted to ICASSP 2026! See you in Barcelona!
Overview
Nord-Parl-TTS is an open TTS dataset for Finnish and Swedish based on speech found in the wild. Using recordings of Nordic parliamentary proceedings, we extract 900 hours of Finnish and 5090 hours of Swedish speech suitable for TTS training. The dataset is built using an adapted version of the Emilia data processing pipeline and includes unified evaluation sets to support model development and benchmarking.
Structure of the Dataset
Train
The training datasets are located under the train folder. Both Finnish and Swedish are stored in the webdataset format.
The structure of the tar file is:
{fi,sv}-000000.tar
- key1.mp3
- key1.json
- key2.mp3
- key2.json
- ...
{fi,sv}-000001.tar
{fi,sv}-000002.tar
...
The keys in the JSON files are slightly different for Finnish and Swedish.
Finnish
An example of the JSON file in the Finnish dataset:
{
"sample_rate": 24000,
"start": START_TIME_IN_THE_RECORDING,
"end": END_TIME_IN_THE_RECORDING,
"dnsmos": DNSMOS P.835 OVL,
"__key__": the filename without extension for mp3 and json,
"text_normalized": Normalized_Transcript_Without_Capital_Letters_And_Punctuations,
"text": Transcript
}
Swedish
An example of the JSON file in the Swedish dataset:
{
"sample_rate": 24000,
"start": START_TIME_IN_THE_RECORDING,
"end": END_TIME_IN_THE_RECORDING,
"dnsmos": DNSMOS P.835 OVL,
"__key__": the filename without extension for mp3 and json,
"text_normalized": Normalized_Transcript_Without_Capital_Letters_And_Punctuations,
"text": Transcript,
"wer": the Word Error Rate between the transcript from Swedish Whisper-large and the given transcript
}
Eval/Test
The evaluation sets are located under the test folder.
For both Finnish and Swedish, there is an eval.lst and an audio.tar.gz.
The eval.lst is organized as:
propmt prompt_duration prompt_transcript target target_duration target_transcript
with a \t in between every two fields.
The audio.tar.gz contains all the prompt and target audio.
Usage
Due to the difference in the JSON file, the Finnish and Swedish training datasets can only be loaded separately.
To load the Finnish training data:
from datasets import load_dataset
finnish_files = "train/finnish/*.tar"
data_fi = load_dataset(
"Aalto-Speech-Synthesis/Nord-Parl-TTS",
streaming=True,
split='fi',
data_files={'fi': finnish_files}
)
print(data_fi)
print(next(iter(data_fi))
For Swedish data:
from datasets import load_dataset
swedish_files = "train/swedish/*.tar"
data_sv = load_dataset(
"Aalto-Speech-Synthesis/Nord-Parl-TTS",
streaming=True,
split='sv',
data_files={'sv': swedish_files}
)
print(data_sv)
print(next(iter(data_sv))
For evaluation sets, they are not supposed to be loaded by the Dataset interface.
Instead, download and decompress them locally.
To download the evaluation sets, you should run
from datasets import load_dataset
eval_data = load_dataset(
"Aalto-Speech-Synthesis/Nord-Parl-TTS",
split='test',
cache_dir="/path/you/wish/to/store/the/evaluation/set"
)
In this way, the evaluation sets will be downloaded to the local directory you wish to use.
You can then decompress the audio.tar.gz and implement your own data interface for evaluation.
Reference
Coming soon
- Downloads last month
- 79
