Seeking AI Training Data Partners

I represent 天津市南开飞凡智序科技有限责任公司, an established provider of AI data and digital-technology services in China. I am expanding our B2B network and am now looking for international AI training-data suppliers or data-labelling companies that can work with us on a long-term, project-based basis. Our most immediate interest is in multi-language speech work: large-scale voice collection, annotated transcriptions and ASR-ready datasets covering a wide range of languages, accents and acoustic environments. If you already curate such corpora or have the capacity to build them, I would like to explore joint delivery models, co-branded offerings or white-label outsourcing. What we bring to the table • A 300 + seat data-ops team experienced in speech, NLP and multimodal annotation • Proven pipelines for data cleaning, segmentation, quality assurance (auto + human) and secure delivery • In-house tools for speaker recruitment, task routing, progress tracking and acceptance testing • The ability to scale quickly inside China for localised collection while meeting global compliance standards What I hope to receive from you 1. Multi-language speech corpora (or a plan to create them) with clear recording specs 2. Time-aligned, language-accurate transcriptions and speaker metadata 3. Your QA methodology and expected word error rate / label accuracy benchmarks 4. Proposed collaboration structure—fixed-price, revenue share or phased milestones (timing is flexible; we have no hard deadline) If your portfolio also includes text, image or multimodal datasets, that could become part of a broader framework agreement, though speech is our immediate priority. Please share a brief capability overview, sample documentation or links to past projects so that we can schedule a technical discussion and, where appropriate, arrange an NDA. I look forward to building a reliable, mutually beneficial partnership that delivers high-quality AI training data to clients worldwide. Budget: USD 30–250