Challenge and purpose: Patients with hematological malignancies are under both physical and psychological pressure due to the impact of the disease and treatment. Chinese healthcare providers are too busy to pay attention to patients’ cancer-related fatigue, sleep, and emotional state. These states often rely on self-reports, making it difficult to assess timely. To address these challenges, we propose a hematologic malignancies clinical corpus 2025 (HMCC 2025) to explore the application of voice analysis in monitoring cancer-related fatigue, sleep quality, and emotional states in these patients. Our research explored the application of speech analysis in clinical nursing practice in order to advance the development of precision nursing care. Methodology: The proposed HMCC involves four stages. (1) The questionnaires consisted of four elements: baseline characteristics, the Chinese version of the Brief Fatigue Inventory (BFI-C), the Pittsburgh Sleep Quality Index (PSQI), and the Hospital Anxiety and Depression Scale (HADS). We use questionnaire speech, scripted speech, and free speech to collect voice from patients with hematologic malignancies. And we record hemoglobin and albumin values through the electronic medical record (EMR) system. (2) We enhance speech quality and usability through noise reduction, silence removal, and signal enhancement. (3) We employ the OpenSMILE to extract 88-dimensional acoustic features, e.g., fundamental frequency characteristics, energy intensity, and spectral structure. (4) We correlate the patients’ physiological and psychological characteristics with the speech characteristics to establish HMCC2025. Results: A total of 87 subjects between the ages of 18 and 88 were enrolled. The BFI-C mean score is 2.31 ± 1.47, the PSQI mean score is 8.24±2.88, the HADS anxiety mean score is 6.83±3.97, and the mean HADS depression score is 6.82±3.69. The mean value of hemoglobin is 109.20 ± 28.48, and the mean value of albumin is 40.91 ± 4.58.