{
  "abstract": "BACKGROUNDArtificial intelligence (AI) medical scribes rely on speech-to-text (STT) models for transcription. Evaluations of STT models in non-English settings remain scarce. We benchmarked ten STT models on medical consultations from Latin American (LatAm) Spanish and assessed whether fine-tuning improves transcription accuracy.\n\nMETHODSTen videos depicting medical consultations. Human transcriptions were the ground truth. Five closed-source models were evaluated: gpt-4o-transcribe, gpt-4o-mini-transcribe, gemini-2.5-pro, Eleven Labs, and Assembly AI; and so were five open-source models: Whisper Large, Whisper Large v3, Whisper Large v3 Turbo, Voxtral Mini 3B, and Canary 1B v2. Whisper Large v3, an open-source model, was fine-tuned. One video was withheld from training. Performance assessed using Word Error Rate (WER), Character Error Rate (CER), BLEU Score, ROUGE-L, BERT Score, and Semantic Similarity on the one withheld video.\n\nRESULTSNone of the fine-tuning iterations outperformed the vanilla Whisper Large v3. When evaluated on the held-out test video, the fine-tuned open-source Whisper Large v3 did not outperform any of the top closed-source models; conversely, relative to other open-source models, the fine-tuned Whisper Large v3 had superior performance across all metrics, such as in BLEU score (63% vs. 58% for the second-best model), BERT Score (89% vs. 86%), and semantic similarity (89% vs. 83%).\n\nCONCLUSIONSWhisper Large v3 and its fine-tuned variant are the best open-source STT models for transcribing medical conversations in LatAm Spanish. These findings provide evidence for developing AI medical scribes tailored to Spanish-speaking LatAm.",
  "abstract_shown": true,
  "affiliations": [
    "Emory University"
  ],
  "article_type": "PUBLISHAHEADOFPRINT",
  "arxiv": null,
  "author_total": 3,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "R. M. Carrillo-Larco"
    },
    {
      "name": "A. Carbajal Serrano"
    },
    {
      "name": "P. S. Condori Pinedo"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/dqpw2zq7te1ea.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.07.14.26358062",
  "fulltext_url": "https://www.medrxiv.org/content/10.64898/2026.07.14.26358062v3.full",
  "id": "dqpw2zq7te1ea",
  "issns": [],
  "landing_url": "https://www.medrxiv.org/content/10.64898/2026.07.14.26358062v3",
  "language": null,
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "medrxiv",
      "public and global health"
    ],
    [
      "openalex",
      "1702|Speech Recognition and Synthesis"
    ]
  ],
  "openalex": "W7168553347",
  "pdf_url": "https://www.medrxiv.org/content/10.64898/2026.07.14.26358062v3.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-07-14",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/dqpw2zq7te1ea.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:20:49Z",
      "native_id": "10.64898/2026.07.14.26358062",
      "source": "medrxiv",
      "url": "https://www.medrxiv.org/content/10.64898/2026.07.14.26358062v3"
    }
  ],
  "subjects": [
    {
      "field": "medicine",
      "origin": "medrxiv:public and global health",
      "subfield": "epidemiology-public-health"
    }
  ],
  "title": "Benchmarking Speech Recognition Models for Medical Consultations in Latin American Spanish: A Comparative Evaluation with Fine-Tuning",
  "title_text": "Benchmarking Speech Recognition Models for Medical Consultations in Latin American Spanish: A Comparative Evaluation with Fine-Tuning",
  "type": "preprint",
  "updated_at": "2026-10-30T12:20:49Z",
  "url": "https://research.draconitic.com/paper/dqpw2zq7te1ea",
  "venue": "medRxiv",
  "versions": []
}