This page covers some of the diagnostics tools and scribing tools in healthcare using generative AI but with specific focus on tools extracting audio data from various sources including sensors, consensual recordings of speech or biological sounds.
The list spans tools across all stages of maturity right from pre-clinical research prototypes to FDA-cleared or CE-marked commercial products. Entries are organised by primary clinical domain. The time-frame for this compilation spans tools launched between January 2013 and December 2025. It does not yet reflect tools launched in 2026.
Respiratory health Includes tools that use cough, breath, or lung sounds for screening, diagnosis or monitoring of respiratory conditions such as asthma, COPD, pneumonia, or tuberculosis.
Tool One-line Description Developer Country Year Audio Input Deployment Regulatory Status Key Validation Swaasa AI-powered cough analysis app for respiratory disease screening Salcit Technologies India 2017 Sound (Cough) App, API (B2B, B2C) India-CDSCO, ISO 13485, ISO 27001, HIPAA; Canada-Therapeutic Products Directorate; US(FDA)-Centre for Devices and Radiological Health; EU- Medical Device Regulation; Australia- TGA Scientific Reports (2023) CoughPro Passive cough monitoring and tracking app for chronic cough management Hyfe Inc. USA 2022 Sound (Cough) App (B2C, B2B) EU- GDPR, Australia- Privacy Act Scientific Reports (2025) ResAppDx Smartphone-based cough diagnostic tool acquired and integrated by Pfizer ResApp Health (acq. Pfizer, 2022) Australia 2015 Sound (Cough) API, telehealth platform (B2B) EU- CE Mark, GDPR; AU- TGA, Privacy Act; Global- ISO 13485, IEC 62304 SMARTCOUGH-C 2 trial, European Respiratory Journal (2021) LungPass Bluetooth stethoscope app for AI lung auscultation at home Healthy Networks OU (ChestPal Ltd) Estonia 2021 Sound (Bluetooth stethoscope for auscultation sounds) App + Device (B2C) EU- CE Mark, GDPR; Global- ISO 13485, IEC 62304, HIPAA European Respiratory Journal (2021) Voicemed.io Multi-modal voice, breath, and cough biomarker platform for respiratory monitoring VoiceMed Luxembourg / Italy 2020 Voice + Sound (Vocal task and breathing sounds) App (B2B, B2C) None (pre-clinical) Chiesi Group pilot (2026)
Mental health Includes voice and speech biomarker tools validated or designed for depression, anxiety, bipolar disorder, PTSD, or general mental wellness screening.
Tool One-line Description Developer Country Year Audio Input Deployment Regulatory Status Key Validation Kintsugi Voice Voice biomarker API for depression and anxiety detection in primary care (now open-sourced) Kintsugi Mindful Wellness, Inc. (ceased commercial ops 2026; open-sourced) USA 2019 Voice (Free form speech) App (B2C), API (B2B) Prior to open-sourcing: HIPAA, GDPR, SOC 2 Type II Annals of Family Medicine (2025) Sonde Health Vocal biomarker SDK for mental health and respiratory fitness screening Sonde Health, Inc. USA 2020 Voice (Speech and phonation) API (B2B) HIPAA, GDPR, SOC 2 Frontiers in Psychiatry (2024) Ellipsis Health AI analysis of call audio for depression and anxiety severity scoring Ellipsis Health USA 2020 Voice (conversational) API, App (B2B) HIPAA, GDPR, SOC 2 Type II, AI Ethics Policy JMIR AI (2025)
Neurological health Includes speech and language analysis tools used for cognitive decline, Parkinson's disease, Alzheimer's, ALS, aphasia or other neurological and neurodegenerative conditions.
Tool One-line Description Developer Country Year Audio Input Deployment Regulatory Status Key Validation TELL App (Toolkit to Examine Lifelike Language) Speech-based neurolinguistic assessment toolkit for cognitive and neurological research Adolfo M. García et al. / Sigmind; Global Brain Health Institute Argentina / International 2023 Voice (Standard speech tasks) Web app (B2B, research) None Behavior Research Methods (2023) Canary Ambient/Speech Enterprise speech analytics for neurological condition detection including Huntington's Canary Speech Inc USA, Utah 2014 Voice (Conversational) Enterprise API HIPAA, GDPR Huntington's Disease study, Beth Israel Deaconess Medical Center Enuncia.ai Mobile speech biomarker app for neurological and communication disorder assessment Enuncia.ai Germany 2022 Voice + Sounds Mobile app GDPR None found Winterlight Labs (now Cambridge Cognition) Speech-based cognitive assessment platform for Alzheimer's and dementia research Winterlight Labs Inc. Canada 2015 Voice (Standard speech tasks) Trials/Research None Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring
Cardiovascular health Includes tools using heart sounds, lung sounds or voice biomarkers associated with cardiac conditions such as arrhythmias, murmurs, heart failure, or hypertension.
Tool One-line Description Developer Country Year Audio Input Deployment Regulatory Status Key Validation Eko Health AI-powered digital stethoscope and ECG platform for cardiac murmur and arrhythmia detection Eko Health USA 2013 Sounds (Heart/lung sounds + ECG) Device + App (B2B, B2C) FDA cleared (murmur, AFib, low EF, EFAST model); EU- CE Mark, UKCA Mark; Global- MDSAP, ISO 13485 TRICORDER study, The Lancet (2025) Vocalis Health Voice biomarker platform for cardiovascular and respiratory remote monitoring Vocalis Health (merger of Beyond Verbal + Healthymize) USA 2019 Voice (Free speech) SDK, web portal (B2B) CE Mark (COVID-19 screening) Mayo Clinic collaboration studies Beyond Verbal Emotion and wellness analytics via voice intonation (merged into Vocalis Health 2019) Beyond Verbal Communication Ltd. (acq. by Vocalis Health, 2019) Israel 2012 Voice (Free form speech) API (B2B); consumer apps Moodie, Empath None Mayo Clinic collaboration studies AiSteth AI Bluetooth stethoscope for cardiac and lung auscultation in low-resource settings Ai Health Highway India Pvt. Ltd. India 2019 Sounds (Heart/lung sounds) Device + App (B2B, B2C) CDSCO, HIPAA NITI Aayog Frontier Tech Hub
Hearing health Includes AI-enhanced hearing aids, audiometric screening apps, sound therapy platforms and tools that analyse ambient sound for hearing loss management or rehabilitation.
Tool One-line Description Developer Country Year Audio Input Deployment Regulatory Status Key Validation Widex SoundSense Learn / MySound Machine learning hearing aid personalisation using real-world ambient sound feedback Widex (now part of WS Audiology) Denmark 2018 Sounds (Ambient sound capture) Hearing aid + App (B2C via audiologist) CE Mark, FDA (hearing aid class) Internal Widex clinical studies Oticon Intent Hearing aid with 4D biometric sensors that adapts to user intent and listening environment Oticon (Demant Group) Denmark 2024 Sounds (Ambient sound + 4D biometric sensors) Hearing aid (B2C via audiologist) CE Mark, FDA (hearing aid class) Oticon Intent white paper (2024) AudioCardio Consumer app delivering personalised sound therapy for hearing health and tinnitus AudioCardio Inc. USA 2019 Sounds App (B2C) None Hearing Tracker coverage
Clinical documentation Includes ambient AI scribes, clinical speech recognition tools and call-audio documentation systems used by clinicians for note generation (these are not diagnostic tools).
Tool One-line Description Developer Country Year Audio Input Deployment Regulatory Status Key Validation Augnito AI clinical speech recognition and documentation platform for healthcare enterprises Scribetech India Healthcare Pvt. Ltd. India / UK 2020 Voice (clinician dictation) App, Desktop, API (B2B, B2C) HIPAA, GDPR, ISO 27001; India- CDSCO Augnito white paper: Enhancing clinician efficiency Heidi Health AI medical scribe that transcribes and structures clinical consultations in real time Heidi Health Australia 2019 Voice (consultation audio) Web, App, API (B2B, B2C) HIPAA, GDPR, NHS, Australia- APP Comparison study, Clinical Endoscopy (2023) IFLYHEALTH China's leading healthcare AI platform using voice interaction and speech recognition for clinical documentation, AI-assisted diagnosis, and chronic disease management Xunfei Healthcare Technology Co. Ltd (subsidiary of iFLYTEK) China 2016 Voice (clinician dictation + patient voice interaction) Enterprise B2B China NMPA; CMMI Level 5 DEV 2.0 No peer-reviewed English paper found Nuance Dragon Medical One Enterprise cloud-based clinical speech recognition used widely across hospital systems Nuance Communications (subsidiary of Microsoft) USA 2013 Voice (clinician dictation) Enterprise API, desktop (B2B) HIPAA, SOC 2, ISO 27001, HITRUST Journal of the American Medical Informatics Association (2024) Tortus Ambient AI consultation recorder and note generator for UK NHS clinical settings Tortus UK 2024 Voice (consultation audio) App (B2B) GDPR, NHS DTAC Pilot programmes with Great Ormond Street Hospital / NHS Trusts Eleos Health AI documentation tool for behavioural health therapy session notes and outcomes Eleos Health USA 2025 Voice (consultation audio) App (B2B) HIPAA Randomised controlled trial, Journal of Medical Internet Research (2023) VoiceOC Automated patient call transcription and documentation for administrative workflows VoiceOC USA 2019 Voice (patient call audio) API, telephony (B2B) None (administrative tool) Allure Medical case study Sunoh.ai Ambient AI scribe for outpatient consultations with EHR integration Sunoh.ai USA 2024 Voice (consultation audio) App, API (B2B, B2C) HIPAA, SOC 2 Type II Esperanza Health Centers case study Lyrebird Health AI clinical note generation from consultation audio for Australian clinicians Lyrebird Health Australia 2022 Voice (consultation audio) App (B2B, B2C) Australia-APP, GDPR No peer-reviewed clinical validation study identified Vitalyzer.in AI assistant that interprets medical report audio for patients in India Vitalyzer India 2025 Voice (medical report narration audio) App (B2C) None No peer-reviewed clinical validation study identified
Foundation models Includes large pre-trained audio or speech models specifically relevant to healthcare (open-source or proprietary).
Model One-line description Developer Year Audio Input Access Regulatory Status Key Papers HeAR (Health Acoustic Representations) Google's large audio foundation model pre-trained on 300M+ health acoustic clips Google Research 2024 Cough, breath, biological sounds Open (Hugging Face, Google Cloud) None (research model) HeAR paper EFAST Eko-developed AI model for full cardiac assessment from digital stethoscope recordings Eko Health 2025 Digital stethoscope recordings Proprietary FDA Cleared TRICORDER study, The Lancet (2025) HuBERT Meta's self-supervised speech representation model trained on unlabelled audio Meta AI 2020 Raw, unlabelled human speech Open Source Non-medical wav2vec 2.0 (Meta Research) Audio-PaLM Google DeepMind multimodal model handling speech, text, and medical records Google DeepMind 2024 Speech, Text, Medical Records Enterprise cloud Research model Google Research CLAP Microsoft's contrastive audio-language model for environment and body sound understanding Microsoft 2023 General environment and body sounds Open Source Non-medical CLAP: Learning Audio Concepts from Natural Language Supervision
Notes Kintsugi shut down commercial operations in 2025 after spending approximately $16 million on FDA regulatory submissions. Its voice biomarker models and scientific methodologies were open-sourced. Beyond Verbal was acquired by and merged into Vocalis Health in December 2019 and no longer operates as an independent product.
References