Speech collection format for underrepresented languages, regional variants, and smaller language communities, structured for ASR evaluation, voice coverage, and multilingual robustness.
Rare and Low-Resource Language Sets are built for teams expanding speech systems into languages with limited available data, inconsistent public coverage, or region-specific variation. The format focuses on structured speech capture, language metadata, transcript usability, and delivery consistency.
Collections can be organized around prompted reads, conversational prompts, domain vocabulary, commands, or buyer-defined evaluation phrases.
Structured collection
Rare and Low-Resource Language Sets are scoped around a specific language target and transcription plan.
Metadata depth
Utterance-level alignment can be prepared as standard.
Delivery-ready package
Includes language match, native speaker status, audio quality reviews, plus dataset manifest, CSV catalog, consent status…
Request sample access
Share your target language, script requirements, speaker criteria, transcript standard, and evaluation goals. We’ll reply with a technical scope for the collection.
Domain vocabulary recordings · Dialect or variant notes · Prompted read speech · Conversational prompt responses
Transcript files · Audio recordings by language target · Language ISO code metadata · Script metadata
Native speaker status · Speaker profile metadata · Prompt category metadata · Recording context metadata
Rare and Low-Resource Language Sets are scoped around a specific language target and transcription plan. The structure prioritizes clean collection instructions, speaker eligibility, language metadata, transcript consistency, and delivery schema alignment.
Language match review · Native speaker status review · Audio quality review · Transcript review · Script consistency review · Consent status review · Delivery approval
Audio files by language · Transcript JSON · Transcript CSV where scoped · Language metadata CSV · Speaker metadata CSV · Dataset manifest JSON · CSV file catalog · QA summary · Consent status metadata · Delivery notes
Multilingual dialogue sets with speaker-labeled transcripts, locale metadata, language-span fields, accent or dialect notes, and aligned transcript structures for multilingual ASR and voice-agent evaluation.
Accent-focused speech sets with speaker profile metadata, region fields, prompt categories, transcript alignment, and QA status fields for ASR robustness and voice-agent coverage.
Multi-speaker English dialogue sets with speaker-labeled transcripts, timestamped turns, accent/session metadata, optional separated-channel audio, and buyer-scoped annotations for ASR, diarization, and voice-agent evaluation.
Bilingual dialogue sets with language-span markers, speaker labels, timestamped transcripts, code-switch metadata, and aligned language fields for multilingual ASR and translation evaluation.
Fully scoped custom data collection with buyer-defined modality, schema, metadata fields, annotation layers, QA criteria, manifest structure, delivery package, and collection rules.
Ambient audio sets with scene labels, acoustic context metadata, speech screening, device fields, noise-level estimates, and QA status for acoustic scene and robustness evaluation.
