Communication methods for deaf and hard of hearing people: what actually works
I've spent the last eight years helping companies set up accessibility workflows, and the thing nobody tells you is that most of the failures come from assumptions, not technology. The tools exist. The budget usually exists. What's missing is understanding how different people within the deaf community actually prefer to communicate, and that list changes depending on age, onset, education, and whether they identify culturally as Deaf or medically as deaf.
Understanding deficientes auditivos ou surdos in practice
The Portuguese term covers a wide spectrum, and treating it as one category is the first mistake. A person who lost hearing at forty after a accident approaches communication differently than a congenitally deaf person raised in a signing household. Both are valid. Both need different support structures. The legal definitions in Brazil and Portugal vary too - Brazil's LBI (Lei Brasileira de Inclusão, Law 13.146/2015) uses "pessoa com deficiência auditiva" while Portugal's legal framework references "surdez" with different accommodations. Know which territory you're operating in before you draft policies. Here's the edge case that still makes me uncomfortable: I once worked with a company that mandated interpreter availability for all deaf employees but never consulted the employees about interpreter preference. One worker used LIBRAS, another used ASL with Portuguese captioning, a third relied primarily on speechreading with hearing aids, and a fourth was oral-only and found interpreters actively harmful to their concentration. The policy was technically compliant and practically useless. We fixed it with individualized communication plans, not blanket rules.
Core communication methods and when they fail
Libras (Língua Brasileira de Sinais) is a complete language with its own grammar, not Portuguese gestured word-by-word. When someone learns "gestos" that map directly to Portuguese syntax, they're not using Libras - they're using a pidgin system that misses nuance and can obscure meaning entirely. Professional interpreters take years of training. Watch for the certification: the CEAD (Certificado de Aptidão como Intérprete) or university-level formation programs carry more weight than weekend workshops. CIPA (Communication Assistance for Interpreters) or direct speech interpretation has a narrow failure window: it breaks down in groups larger than four people, in rooms with poor lighting where the interpreter's face isn't visible, or when multiple speakers talk over each other. I've seen meeting recordings where the deaf participant missed the actual discussion because three people were simultaneously conversing and the interpreter couldn't keep pace. The workaround is simple but rarely enforced: one speaker at a time, face visible, pause between contributions.
Writing and text-based communication sounds obvious but is inconsistently applied. Email works. Chat works. But the assumption that everyone who is deaf reads at the same level as a hearing peer is wrong. Some congenitally deaf people have delayed language acquisition because they missed the critical early exposure window. A person who learned Libras as an adult may read Portuguese better than they sign it, or vice versa. Don't assume literacy level from hearing status. Captioning and transcription have improved dramatically with AI, but accuracy drops below 85 percent on technical content, heavy accents, or overlapping speech. Real-time captioning (CART) is still the gold standard for live events. I once attended a conference where the provider used automatic speech recognition instead of a human captioner - the captions were readable but frequently wrong on technical terms, and a deaf attendee missed a critical safety announcement because the algorithm didn't recognize the jargon. Human captioners cost more but prevent embarrassment and, occasionally, physical harm.
Assistive technology: what's worth buying
Assistive listening devices like FM systems and loop systems exist, but installation is the bottleneck. A loop system requires a trained technician to configure the hearing loop around the room's geometry. I've seen "accessible" venues where the loop was installed incorrectly and only worked within a two-meter radius of the receiver. Test before you certify. Same for FM systems - the range depends on venue size and interference from other electronics. Hearing aids and cochlear implants are personal devices with individual fit curves. A cochlear implant user may hear better in quiet one-on-one settings but struggle in noise because the processing prioritizes clarity over ambient sound. Don't assume a hearing aid equals "hears like a hearing person." The device amplifies; it doesn't restore normal hearing. Expect to ask the person what works for them rather than guessing.
👉 Clique no botão abaixo para saber mais sobre o assunto!
Video relay service (VRS) and video remote interpreting (VRI) are standard now, but connection quality matters more than people realize. A frozen video feed during a sign language conversation is worse than no feed at all - you lose the visual grammar entirely. Latency above 300ms makes real-time conversation impossible. Test your internet before booking VRI for important meetings. For anything legal, medical, or contractual, in-person interpreters remain superior. VRI has its place for quick questions, but not for informed consent discussions.
Workplace and institutional adjustments that don't cost much
The easiest wins are procedural. Front-row seating costs nothing and solves half the visibility problems. Written agendas sent before meetings let people prepare and follow along without relying solely on real-time interpretation. Email follow-ups after verbal discussions create a paper trail that benefits everyone, not just deaf participants. Visual alert systems for fire alarms, doorbells, and announcements are required by law in many jurisdictions but notoriously poorly implemented. I audited a building where the visual strobes were installed but positioned behind pillars, so the warning light never reached the person's field of view. Compliance checked the box but didn't check the sightlines. Walk the space yourself or hire someone who will.
Accessible video content means both captioning and, ideally, Libras insertion for Brazilian audiences. Auto-captions from YouTube or other platforms are improving but still miss context. For internal training videos, budget for professional captioning - it takes about 20 to 30 minutes per video hour but prevents the need for rework when captions contain errors that change meaning. The cost is roughly R$150 to R$400 per hour depending on the provider.
When these methods don't work
No single approach covers everyone. A deaf person who is also blind requires (tactile signing) or braille captioning, and very few institutions have staff trained for that combination. A deaf migrant who speaks Arabic and knows only Arabic Sign Language won't benefit from a Libras interpreter. These are edge cases but they happen, and the workaround is individual consultation rather than system assumptions. Real-time captioning fails when the speaker uses rapid colloquial speech, heavy regional dialects, or technical terminology outside the captioner's expertise. The mitigation is pre-briefing: send speaker notes to the captioner or interpreter before the event so they can prepare specialized vocabulary. This cuts error rates significantly.
And the hardest truth: some communication gaps can't be fully resolved through technology or policy alone. A congenitally deaf person raised in a non-signing family may have limited language access regardless of accommodations available. This is why early intervention and family education matter more than workplace adjustments, even though the latter is easier to legislate. If you're starting from zero, the best first step isn't buying equipment. It's talking to deaf people in your organization or community and asking what they actually need. The answer will surprise you, and it will be different for every person you ask.