Most tools don't fail Nepali by refusing it — they fail it by mistaking it for Hindi, then flattening the social levels that carry the meaning. Nepali is written in the same Devanagari script as Hindi, so an auto-detector reads the familiar letters and labels the audio "Hindi," then renders the wrong everyday words. But the deeper miss is structural: Nepali encodes three levels of social respect in both its pronouns and its verb endings — and a flat translator collapses all of them into one toneless "you." Add the migrant-and-diaspora reality of who speaks Nepali on these calls, and the Kathmandu IT-and-NGO habit of mixing Nepali with English, and "supports Nepali" on a feature list tells you almost nothing. Here's what actually decides whether a Nepali meeting comes back usable.
Same script as Hindi, a different language
Nepali and Hindi both use Devanagari, and that shared alphabet is exactly what fools an auto-detector — it recognizes the letters, guesses "Hindi," and writes Hindi vocabulary into a Nepali conversation. But the languages diverge on the words people say most. "I" is ma in Nepali, not Hindi's main; "is" is ho or cha (Nepali distinguishes identity from existence — yo X ho, "this is X," versus X cha, "X exists/there is"), where Hindi leans on hai; "and" is ra, not aur; "but" is tara, not lekin. A model that defaults to Hindi outputs text that looks correct character-by-character and is wrong word-by-word — and the ho/cha distinction in particular is a meaning split, not a spelling preference, so a summary built on the wrong copula can invert what someone actually claimed.
Three honorific levels, in the pronouns and the verbs
This is the part flat translators erase entirely. Nepali grades respect into three levels, and the level shows up twice — in the pronoun you choose for "you," and in how the verb conjugates to agree with it. High respect is tapāī̃; mid/familiar is timī; low/intimate is ta. And "you do" is a different verb form at each level: tapāī̃ garnuhuncha (high), timī garchau (mid), ta garchas (low). Same action, three encodings of the relationship between the people on the call — manager to report, peer to peer, close friend to friend. A translator that knows only English has one word, "you," and one verb, "do," so it flattens garnuhuncha, garchau, and garchas into the same flat line — and the social information that a native listener hears immediately is simply gone from the transcript. On a workforce call where a senior addresses staff with tapāī̃ and that respect is the point, flattening it isn't a cosmetic loss; it's the meeting losing its register.
Who is actually on the call: a workforce spread across continents
Nepali isn't a single-city language on these calls. Nepali speakers are spread across the Gulf, Malaysia, the UK, and distributed teams worldwide as a large migrant and diaspora workforce — which means a "Nepali meeting" is often a mixed-language room: a Nepali-speaking crew or staff on one side, an English-speaking or other-language coordinator on the other, sometimes both in the same call. That's exactly the situation where one shared, flattened caption track fails everyone: the Nepali speakers need clean Nepali with the honorific level intact, and the coordinator needs clean English, from the same spoken words, at the same time. Per-person captions — each participant reading their own language — are not a nice-to-have here; they're the whole point of running a mixed-language meeting at all. (For why this pattern repeats across distributed teams, see real-time translation for remote teams.)
Nenglish is the Kathmandu working register
In Kathmandu IT and NGO work, the working register isn't textbook Nepali — it's Nenglish: a Nepali frame with English nouns and verbs dropped in, often with Nepali endings attached. "Tyo feature aglo sprint mā deploy garnu cha" is one ordinary sentence — English content words, Nepali grammar, Nepali verb. A tool that detects "Nepali" may leave the English untranslated; one that detects "Hindi" mangles both. Each reader needs a complete sentence rebuilt in their own language, not a half-translated line with the English left dangling. (The same code-mixing pattern shows up across the region — see Hindi ↔ English meeting translation for the Hindi-side version.)
How to do it with Sageio
- Add
bot@sageio.netto your Google Meet calendar invite. It joins on its own — no extension, nothing to install. - Each participant picks their caption language. The Nepali-speaking crew reads clean Nepali, a coordinator elsewhere reads clean English — both from the same spoken Nepali, at the same time. (Sageio translates into 20+ languages.)
- Everyone speaks naturally — Nepali, the honorific levels, Nenglish, all of it. Translated captions appear in about two seconds.
- Afterward, a searchable transcript and an AI summary arrive within about five minutes, shared at the host's discretion.
(Today this runs on Google Meet; Zoom and Microsoft Teams support is coming soon.)
How to test any tool in five minutes
Say a plain Nepali sentence that uses the giveaway words — ma, ho/cha, ra, tara — where a Hindi-default model would write main, hai, aur, lekin instead, and check whether the captions come back in Nepali or in Hindi-flavoured Devanagari. Then say the same action at two honorific levels — tapāī̃ garnuhuncha and timī garchau — and see whether the translation preserves any sense of the register shift or flattens both into one toneless "you do." Finally, say a normal Nenglish line ("yo task malāī next week samma finish garnu cha" — "I have to finish this task by next week") and see whether it keeps the English words whole while rendering the Nepali correctly. If it labels you "Hindi," collapses the honorifics, or garbles the mix, the tool wasn't built for spoken Nepali.
Is it private?
For anything that joins your meetings: Sageio doesn't use your meeting content to train AI models, and its AI vendors are contractually restricted from doing the same. Audio is processed in memory and discarded — only the text transcript and summary are kept, encrypted, in the region you choose (US, EU, or APAC). Enterprise customers can self-host the entire stack.
Frequently asked questions
Why does a tool label Nepali as "Hindi"? Nepali and Hindi share the Devanagari script, so an auto-detector that reads the alphabet recognizes the familiar letters and guesses "Hindi." It then writes Hindi vocabulary into a Nepali conversation — text that looks correct character-by-character but reads wrong to a Nepali speaker, because the languages differ on core words like ma vs main ("I"), ra vs aur ("and"), and the ho/cha copulas Hindi handles with hai.
What are Nepali honorific levels, and why do they matter for translation? Nepali grades respect into three levels — high tapāī̃, mid timī, low ta — and the level is encoded in both the pronoun and the verb ending: "you do" is garnuhuncha, garchau, or garchas depending on the relationship. A translator that maps all three onto a single English "you do" erases the social information a native listener registers without thinking, so the transcript loses the register the meeting was conducted in.
Why do per-person caption languages matter for Nepali calls? Nepali speakers are spread across the Gulf, Malaysia, the UK, and distributed teams worldwide, so a Nepali meeting is usually a mixed-language room. The Nepali-speaking side needs clean Nepali and a coordinator needs clean English — from the same spoken words, at the same time. Each participant picking their own caption language is what makes that possible.
What is Nenglish? Nepali grammar with English nouns and verbs mixed in, often carrying Nepali endings — "deploy garnu cha." It's the normal working register in Kathmandu IT and NGO work. Tools that assume one language per sentence translate only half; correct handling rebuilds a full sentence in each target language.
What does it cost to try? Every plan starts with a free 60-minute trial, no credit card required. After that, Professional is $49/month and Teams is $99 per seat/month (annual billing includes 2 months free); Enterprise is custom-priced.
If your team works in Nepali, the honest test is whether a native speaker reads the live captions and transcript and hears the actual meeting — in Nepali, not Hindi-flavoured Devanagari, with the honorific level intact and the Nenglish kept whole. Add the bot to your next call and let them judge.