Hugging Face's logo

Datasets:
atenax
/
kmr_tts

Tasks:
Text-to-Speech
Modalities:
Audio
Languages:
Kurdish
Size:
10K<n<100K
Tags:
audio
speech
kurdish
kurmanji
northern-kurdish
piper
License:
Dataset card Files
xet
 Community
Dataset Viewer
The dataset viewer is not available for this dataset.
Cannot get the config names for the dataset.
Error code:   ConfigNamesError
Exception:    BadZipFile
Message:      File is not a zip file
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/dataset/config_names.py", line 67, in compute_config_names_response
                  config_names = get_dataset_config_names(
                      path=dataset,
                      token=hf_token,
                  )
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 161, in get_dataset_config_names
                  dataset_module = dataset_module_factory(
                      path,
                  ...<4 lines>...
                      **download_kwargs,
                  )
                File "/usr/local/lib/python3.14/site-packages/datasets/load.py", line 1215, in dataset_module_factory
                  raise e1 from None
                File "/usr/local/lib/python3.14/site-packages/datasets/load.py", line 1190, in dataset_module_factory
                  ).get_module()
                    ~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/load.py", line 655, in get_module
                  module_name, default_builder_kwargs = infer_module_for_data_files(
                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      data_files=data_files,
                      ^^^^^^^^^^^^^^^^^^^^^^
                      path=self.name,
                      ^^^^^^^^^^^^^^^
                      download_config=self.download_config,
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/load.py", line 309, in infer_module_for_data_files
                  split: infer_module_for_data_files_list(data_files_list, download_config=download_config)
                         ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/load.py", line 253, in infer_module_for_data_files_list
                  return infer_module_for_data_files_list_in_archives(data_files_list, download_config=download_config)
                File "/usr/local/lib/python3.14/site-packages/datasets/load.py", line 280, in infer_module_for_data_files_list_in_archives
                  f.split("::")[0] for f in xglob(extracted, recursive=True, download_config=download_config)
                                            ~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/utils/file_utils.py", line 1057, in xglob
                  fs, *_ = url_to_fs(urlpath, **storage_options)
                           ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/fsspec/core.py", line 408, in url_to_fs
                  fs = filesystem(protocol, **inkwargs)
                File "/usr/local/lib/python3.14/site-packages/fsspec/registry.py", line 303, in filesystem
                  return cls(**storage_options)
                File "/usr/local/lib/python3.14/site-packages/fsspec/spec.py", line 81, in __call__
                  obj = super().__call__(*args, **kwargs)
                File "/usr/local/lib/python3.14/site-packages/fsspec/implementations/zip.py", line 62, in __init__
                  self.zip = zipfile.ZipFile(
                             ~~~~~~~~~~~~~~~^
                      self.fo,
                      ^^^^^^^^
                  ...<3 lines>...
                      compresslevel=compresslevel,
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/usr/local/lib/python3.14/zipfile/__init__.py", line 1472, in __init__
                  self._RealGetContents()
                  ~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/zipfile/__init__.py", line 1539, in _RealGetContents
                  raise BadZipFile("File is not a zip file")
              zipfile.BadZipFile: File is not a zip file

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

KMR TTS — Kurmanji Speech Corpus

KMR TTS is a processed Kurmanji (Northern Kurdish) speech corpus prepared for text-to-speech training. It was used to fine-tune the multi-speaker atenax/kurmanci-tts-piper model with the Piper training pipeline.

Dataset summary

Property Value
Language Kurmanji / Northern Kurdish (ku)
Modality Speech audio with text transcriptions
Primary task Text-to-speech
Prepared training records 56,499
Speaker identities 90
Packaged size 27.1 GB
Distribution dataset.zip

The prepared metadata uses pipe-separated rows in the form:

audio_id|anonymized_speaker_id|transcription

The local Piper preparation output also includes JSONL records with audio, phoneme, and speaker mappings suitable for training.

Source and preparation

The corpus was prepared from Kurmanji speech data from Mozilla Common Voice 24.0. Preparation included:

  1. Selecting Kurmanji speech and transcription records.
  2. Normalizing text while retaining Kurmanji characters such as ç, ê, î, ş, and û.
  3. Filtering unusable or missing samples.
  4. Converting audio for a consistent speech-training workflow.
  5. Building Piper metadata and anonymized speaker mappings.

The speaker values are opaque hashes used to group utterances; they are not public-facing names.

Download and use

Download the packaged corpus with the Hugging Face CLI:

hf download atenax/kmr_tts dataset.zip \
  --repo-type dataset \
  --local-dir ./kmr_tts

Then extract dataset.zip and inspect the included metadata before training. The current release is distributed as a prepared archive rather than a Hub Dataset Viewer configuration.

Intended use

This dataset is intended for:

  • Kurmanji text-to-speech research.
  • Speech preprocessing and phonemization experiments.
  • Training or evaluating open-source speech synthesis systems.
  • Work on language technology for underrepresented languages.

Limitations and responsible use

  • Transcriptions and speaker labels may contain upstream or preprocessing errors; perform task-specific validation before training.
  • Speaker coverage, dialect, recording conditions, age, and gender are not guaranteed to be balanced.
  • Do not attempt to reverse or expose speaker identities from anonymized IDs.
  • Models trained on this data can reproduce biases and artifacts in the source recordings. Review generated speech before deployment.

Related resources

Citation

When using this release, cite the upstream Mozilla Common Voice dataset and link to this dataset repository so others can identify the exact processed artifact.
Downloads last month
9
Total file size:
27.1 GB

Models trained or fine-tuned on atenax/kmr_tts

Space using atenax/kmr_tts 1