ARXIV:2606.03504 · SPEECH RECOGNITION · SUBMITTED 03 JUN · 20:32 UTC · FRESHNESS FRESH

VerifiedSource: PDF linkedVerifiedPaperPack: citation fields availablePartialProof: partial proof status

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

Muhammad Ali · arXiv

A fine-tuned Whisper ASR system and speech corpus for the Balti language, enabling transcription with a public demo and model.

Ship in 2-4 weeks›Score8.0Evidence partial

Opportunity summary

Pain A fine-tuned Whisper ASR system and speech corpus for the Balti language, enabling transcription with a public demo and model.

Evidence 0 refs | 7 sources | 67% coverage

Blocker Evidence partial

Open Build Read PDF Signal Canvas Track

PROBLEM

A fine-tuned Whisper ASR system and speech corpus for the Balti language, enabling transcription with a public demo and model. The corpus contains 10,060 validated utterances in native Nastaliq script, derived from Mozilla Common…

METHOD

Full 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. We fine-tune OpenAI Whisper-small on this corpus and report a Word Error Rate (WER) of 30.07% on a held-out validation set of 538 utterances, down from a measured zero-shot baseline of 182.18% for Whisper-small on Balti. The dataset, fine-tuned model, and a live transcription demo are publicly available on HuggingFace.

RESULT

ScienceToStartup currently rates this 8.0/10 on the public viability pass. The dataset, fine-tuned model, and a live transcription demo are publicly available on HuggingFace. A public repository is linked, so build verification can inspect…

WHY NOW

Speech Recognition moved forward this cycle; last verified June 2026. Public score 8.0/10. Implementation evidence is present through a linked repository.

Continue into Read for claims, analysis, references, and neighboring papers.

Opportunity summary

Score8.0

PainA fine-tuned Whisper ASR system and speech corpus for the Balti language, enabling transcription with a public demo and model.

Evidence0 refs | 7 sources | 67% coverage

Blockerno shell-level blocker reported

Analysis summary

A fine-tuned Whisper ASR system and speech corpus for the Balti language, enabling transcription with a public demo and model.

VerifiedSource: PDF linkedVerifiedPaperPack: citation fields availablePartialProof: partial proof status

Competitive landscape

A fine-tuned Whisper ASR system and speech corpus for the Balti language, enabling transcription with a public demo and model.

Segment

Speech Recognition

Adoption evidence

Public code linked for build inspection

Commercial read

8.0/10 public viability

Direct

not classified

Adjacent

not classified

Substitute

not classified

Unknown

not classified

{ "contract_version": "paper-r2", "paper_id": "691949ca-b3a9-4245-a451-692fc02d5e4e", "arxiv_id": "2606.03504", "canonical_route": "/paper/baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language", "active_tab": "synced from current hash by the drawer client", "selected_artifact": "baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language", "endpoints": { "paper_pack": "/api/v1/paper/baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language/paper-pack", "build_passport": "/api/v1/paper/baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language/build-passport", "mcp_resource": "sciencetostartup://surfaces/paper-workspace" } }

{ "surface": "paper", "mode": "paper", "query": "BaltiVoice: A Speech Corpus and Fine-tuned Whisper ASR System for the Balti Language", "normalized_query": "2606.03504", "route": "/paper/baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language", "paper_ref": "baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language", "topic_slug": null, "benchmark_ref": null, "dataset_ref": null }

{ "@context": "https://schema.org", "@graph": [ { "@type": "WebPage", "@id": "https://sciencetostartup.com/paper/baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language#webpage", "url": "https://sciencetostartup.com/paper/baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language", "name": "BaltiVoice: A Speech Corpus and Fine-tuned Whisper ASR System for the Balti Language", "description": "A fine-tuned Whisper ASR system and speech corpus for the Balti language, enabling transcription with a public demo and model.", "isPartOf": { "@id": "https://sciencetostartup.com/#website" } }, { "@type": "ScholarlyArticle", "@id": "https://sciencetostartup.com/paper/baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language#scholarlyArticle", "headline": "BaltiVoice: A Speech Corpus and Fine-tuned Whisper ASR System for the Balti Language", "description": "A fine-tuned Whisper ASR system and speech corpus for the Balti language, enabling transcription with a public demo and model.", "url": "https://sciencetostartup.com/paper/baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language", "sameAs": "https://arxiv.org/abs/2606.03504", "identifier": { "@type": "PropertyValue", "propertyID": "arXiv", "value": "2606.03504" }, "isAccessibleForFree": true, "isPartOf": { "@id": "https://sciencetostartup.com/#website" }, "datePublished": "2026-06-02T11:23:49.000Z", "author": [ { "@type": "Person", "name": "Muhammad Ali" } ], "codeRepository": "https://github.com/mohdali-dev/BaltiVoice-ASR", "additionalProperty": [ { "@type": "PropertyValue", "propertyID": "viabilityScore", "value": 8 }, { "@type": "PropertyValue", "propertyID": "researchDomain", "value": "Speech Recognition" }, { "@type": "PropertyValue", "propertyID": "commercialReadiness", "value": "code, repo url" } ] }, { "@type": "SoftwareSourceCode", "@id": "https://sciencetostartup.com/paper/baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language#software", "name": "BaltiVoice: A Speech Corpus and Fine-tuned Whisper ASR System for the Balti Language - Source Code", "description": "A fine-tuned Whisper ASR system and speech corpus for the Balti language, enabling transcription with a public demo and model.", "codeRepository": "https://github.com/mohdali-dev/BaltiVoice-ASR", "url": "https://github.com/mohdali-dev/BaltiVoice-ASR" }, { "@type": "BreadcrumbList", "itemListElement": [ { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://sciencetostartup.com" }, { "@type": "ListItem", "position": 2, "name": "Speech Recognition", "item": "https://sciencetostartup.com/topics" }, { "@type": "ListItem", "position": 3, "name": "BaltiVoice: A Speech Corpus and Fine-tuned Whisper ASR Syste", "item": "https://sciencetostartup.com/paper/baltivoice-a-speech-corpus-and-fine-tuned-whisper-asr-system-for-the-balti-language" } ] } ] }

Competitive landscape

A fine-tuned Whisper ASR system and speech corpus for the Balti language, enabling transcription with a public demo and model.

Segment

Speech Recognition

Adoption evidence

Public code linked for build inspection

Commercial read

8.0/10 public viability

Direct

not classified

Adjacent

not classified

Substitute

not classified

Unknown

not classified

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

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

Claim map

Constellation map

Competitive landscape

Buzz

PDF

REFERENCES

Related Papers

Related Resources

Subscribe to the weekly brief

Build artifacts

Brief

Experiment plan

Validation checklist

Scientific founder

Translational engineer

Domain operator

GTM lead

Regulatory/clinical advisor

Timeline

Claim map

Constellation map

Competitive landscape

Buzz

PDF

REFERENCES

Related Papers

Related Resources

Subscribe to the weekly brief

Build artifacts

Brief

Experiment plan

Validation checklist

Scientific founder

Translational engineer

Domain operator

GTM lead

Regulatory/clinical advisor

Timeline