BaltiVoice: A Speech Corpus and Fine-tuned Whisper ASR System for the Balti Language

• URL: http://arxiv.org/abs/2606.03504v2
• Date: Wed, 10 Jun 2026 09:34:48 GMT
• Title: BaltiVoice: A Speech Corpus and Fine-tuned Whisper ASR System for the Balti Language
• Abstract summary: We present BaltiVoice, a 16.8-hour read-speech corpus for Balti (ISO 639-3: bft), a Tibetic language spoken in Gilgit-Baltistan, Pakistan.<n>The corpus contains 10,060 validated utterances in native Nastaliq script, derived from Mozilla Common Voice recordings.<n>Fine-tuning OpenAI Whisper-small yields a Word Error Rate (WER) of 26.74% and a Character Error Rate (CER) of 8.67% on a 538-utterance speaker-disjoint validation set.
• Score: 0.20771014670599888
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We present BaltiVoice, a 16.8-hour read-speech corpus for Balti (ISO 639-3: bft), a Tibetic language spoken in Gilgit-Baltistan, Pakistan, with no prior publicly available ASR resources. The corpus contains 10,060 validated utterances in native Nastaliq script, derived from Mozilla Common Voice recordings. Fine-tuning OpenAI Whisper-small yields a Word Error Rate (WER) of 26.74% and a Character Error Rate (CER) of 8.67% on a 538-utterance speaker-disjoint validation set, down from a zero-shot baseline of 159.19% WER and 152.52% CER. A Whisper-base fine-tuned on the same data achieves 44.54% WER and 15.61% CER, confirming that model capacity matters for this low-resource setting. The dataset, fine-tuned model, and a live transcription demo are publicly available on HuggingFace.

Related papers

• Robust Assamese Speech Recognition through Controlled Fine-Tuning of Whisper Models [1.4095958360608893]
  This paper presents a controlled, fine-tuned Whisper-based Assamese ASR system trained on the Mozilla Common Voice 24.0-Assamese corpus.<n>A hardware-aware optimized training pipeline is implemented for resource-constrained environments.
  arXiv  Detail & Related papers  (2026-07-19T09:54:52Z)
• From Speech to Text Corpora: Evaluating ASR-Based Data Acquisition for Low-Resource Fongbe and Hausa [11.053174014226078]
  We investigate whether ASR pipelines can extend text resources for two typologically distinct West African languages.<n>We fine-tune MMS-300M on a curated 12.3-hour Fongbe dataset, achieving 9.48% WER on the ALFFA benchmark.<n>We catalog 1,553 YouTube videos (236 hours) and process a subset of 424 videos (45.49 hours) selected to balance domain diversity with available computational resources.<n>Human evaluation on 50 randomly sampled segments per language shows mean quality scores of 57.4/100 for Hausa and 36.5/100 for Fongbe.
  arXiv  Detail & Related papers  (2026-06-20T23:51:55Z)
• PashtoTTS-Bench: automated screening for low-resource non-Latin-script text-to-speech [0.0]
  Text-to-speech evaluation for low-resource non-Latin-script languages can fail when it relies on a single ASR round-trip word error rate (WER)<n>We introduce INSV, a reporting framework that separates these cases.<n>We instantiate INSV-A as PashtoTTS-Bench, a dated benchmark for Pashto TTS.
  arXiv  Detail & Related papers  (2026-05-26T13:03:20Z)
• BlasBench: An Open Benchmark for Irish Speech Recognition [0.0]
  Existing benchmarks include Irish among dozens of languages but apply no Irish-aware text normalisation.<n>We introduce BlasBench, an open evaluation harness that provides a standalone Irish-aware normaliser preserving fadas, lenition, and eclipsis.<n>We pilot this by benchmarking 12 systems across four architecture families on Common Voice ga-IE and FLEURS ga-IE.
  arXiv  Detail & Related papers  (2026-04-12T17:17:54Z)
• Benchmarking Multilingual Speech Models on Pashto: Zero-Shot ASR, Script Failure, and Cross-Domain Evaluation [0.0]
  Pashto is spoken by approximately 60--80 million people but has no published benchmarks for multilingual automatic speech recognition (ASR) on any shared public test set.<n>This paper reports the first reproducible multi-model evaluation on public Pashto data, covering zero-shot ASR, script-level failure, and cross-domain evaluation of fine-tuned models.
  arXiv  Detail & Related papers  (2026-04-06T11:23:42Z)
• Pashto Common Voice: Building the First Open Speech Corpus for a 60-Million-Speaker Low-Resource Language [0.0]
  We present the Pashto Common Voice corpus -- the first large-scale, openly licensed speech resource for Pashto.<n>Through a community effort spanning 2022-2025, the corpus grew from 1.5 hours and 5 contributors to 147 total hours and 1,483 unique speakers.<n>Speaker participation increased approximately 108-fold between CV17 and CV18, coinciding with a VOA Pashto broadcast campaign.
  arXiv  Detail & Related papers  (2026-03-27T22:22:03Z)
• Qwen3-ASR Technical Report [71.87071808763484]
  We introduce Qwen3-ASR family, which includes two powerful all-in-one speech recognition models and a novel non-autoregressive speech forced alignment model.<n>Qwen3-ASR-1.7B and Qwen3-ASR-0.6B are ASR models that support language identification and ASR for 52 languages and dialects.
  arXiv  Detail & Related papers  (2026-01-29T06:58:13Z)
• The ML-SUPERB 2.0 Challenge: Towards Inclusive ASR Benchmarking for All Language Varieties [107.57160730151975]
  We construct a new test suite that consists of data from 200+ languages, accents, and dialects to evaluate SOTA multilingual speech models.<n>The best-performing submission achieved an absolute improvement in LID accuracy of 23% and a reduction in CER of 18%.<n>On accented and dialectal data, the best submission obtained 30.2% lower CER and 15.7% higher LID accuracy.
  arXiv  Detail & Related papers  (2025-09-08T18:42:36Z)
• Exploring SSL Discrete Speech Features for Zipformer-based Contextual ASR [74.38242498079627]
  Self-supervised learning (SSL) based discrete speech representations are highly compact and domain adaptable.<n>In this paper, SSL discrete speech features extracted from WavLM models are used as additional cross-utterance acoustic context features in Zipformer-Transducer ASR systems.
  arXiv  Detail & Related papers  (2024-09-13T13:01:09Z)
• A Large Dataset of Spontaneous Speech with the Accent Spoken in São Paulo for Automatic Speech Recognition Evaluation [0.8929537856623965]
  We present a freely available spontaneous speech corpus for the Brazilian Portuguese language. We report preliminary automatic speech recognition (ASR) results, using both the Wav2Vec2-XLSR-53 and Distil-Whisper models. To the best of our knowledge, this is the first large Paulistano accented spontaneous speech corpus dedicated to the ASR task in Portuguese.
  arXiv  Detail & Related papers  (2024-09-10T21:45:06Z)
• ÌròyìnSpeech: A multi-purpose Yorùbá Speech Corpus [7.97238074132292]
  IroyinSpeech is a new corpus influenced by the desire to increase the amount of high quality, contemporary Yorub'a speech data. We curated about 23000 text sentences from news and creative writing domains with the open license CC-BY-4.0.
  arXiv  Detail & Related papers  (2023-07-29T20:42:50Z)
• Voicebox: Text-Guided Multilingual Universal Speech Generation at Scale [58.46845567087977]
  Voicebox is the most versatile text-guided generative model for speech at scale. It can be used for mono or cross-lingual zero-shot text-to-speech synthesis, noise removal, content editing, style conversion, and diverse sample generation. It outperforms the state-of-the-art zero-shot TTS model VALL-E on both intelligibility (5.9% vs 1.9% word error rates) and audio similarity (0.580 vs 0.681) while being up to 20 times faster.
  arXiv  Detail & Related papers  (2023-06-23T16:23:24Z)
• ASR2K: Speech Recognition for Around 2000 Languages without Audio [100.41158814934802]
  We present a speech recognition pipeline that does not require any audio for the target language. Our pipeline consists of three components: acoustic, pronunciation, and language models. We build speech recognition for 1909 languages by combining it with Crubadan: a large endangered languages n-gram database.
  arXiv  Detail & Related papers  (2022-09-06T22:48:29Z)
• XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale [48.0390317915984]
  XLS-R is a large-scale model for cross-lingual speech representation learning based on wav2vec 2.0. We train models with up to 2B parameters on nearly half a million hours of publicly available speech audio in 128 languages.
  arXiv  Detail & Related papers  (2021-11-17T18:49:42Z)
• A Crowdsourced Open-Source Kazakh Speech Corpus and Initial Speech Recognition Baseline [4.521450956414864]
  The Kazakh speech corpus (KSC) contains around 332 hours of transcribed audio comprising over 153,000 utterances spoken by participants from different regions and age groups. The KSC is the largest publicly available database developed to advance various Kazakh speech and language processing applications.
  arXiv  Detail & Related papers  (2020-09-22T05:57:15Z)
• Unsupervised Cross-lingual Representation Learning for Speech Recognition [63.85924123692923]
  XLSR learns cross-lingual speech representations by pretraining a single model from the raw waveform of speech in multiple languages. We build on wav2vec 2.0 which is trained by solving a contrastive task over masked latent speech representations. Experiments show that cross-lingual pretraining significantly outperforms monolingual pretraining.
  arXiv  Detail & Related papers  (2020-06-24T18:25:05Z)
• CoVoST: A Diverse Multilingual Speech-To-Text Translation Corpus [57.641761472372814]
  CoVoST is a multilingual speech-to-text translation corpus from 11 languages into English. It diversified with over 11,000 speakers and over 60 accents. CoVoST is released under CC0 license and free to use.
  arXiv  Detail & Related papers  (2020-02-04T14:35:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.