Gemini Live
Hacker ApprovedRanked 2 of 6 voice AI models by recommenders
People recommend it for fluent, low-latency spoken conversation across many languages, and for tutoring and practice. Critics say it sometimes loses context and misidentifies some languages.
People recommend it for fluent, low-latency spoken conversation across many languages, and for tutoring and practice. Critics say it sometimes loses context and misidentifies some languages.
73 people gave an opinion
- recommend it
- 50
- criticize it
- 23
- approval share
- 68%
Recommended in 9 of 13 months
- November 2025: not recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: not recommended
- March 2026: not recommended
- April 2026: recommended
- May 2026: recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: recommended
- September 2026: recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it carries the badge
- 50 people recommend itNeeds 10 or more, met.
- 2.2 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 9 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Strong multilingual and dialect support
“I was also surprised at how good Gemini Live mode is in regional Indian languages”
Comment from 11 Dec 2025 in “GPT-5.2” “what impressed me was its ability to recognize the subtle dialects of shona being spoken”
Comment from 16 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking” “its working fantasticly, switching between german/english with no effort”
Comment from 25 Oct 2026 in “No signs of AI payoff outside the tech sector” - Comment from 16 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking”
- Comment from 15 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking”
- Comment from 16 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking”
Natural, human-like conversation
“It really does feel like talking to a real person.”
Comment from 15 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking” “the live audio/video back and forth we had with the AI was almost magical”
Comment from 5 Jun 2026 in “What was your "Oh Shit" moment with GenAI?” “It kinda was fun, like a very patient professor stand right besides you. It was the one of the best math learning experience I've ever had”
Comment from 4 Jun 2026 in “Failing grades soar with AI usage, dwindling math skills in Berkeley CS classes” - Comment from 9 Jul 2026 in “GPT‑Live”
Handles interruptions well
“it really works surprisingly well, especially the interruption”
Comment from 8 Jul 2026 in “GPT‑Live” “when you interrupt it, it shuts up immediately”
Comment from 15 Sep 2026 in “Gemini 3.8 Live Extended Thinking audio-to-audio” Good for language practice
“Benchmarks don't capture the magic of having a 24/7 native-speaking conversational partner in a rare language.”
Comment from 16 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking” “I've been using Gemini Live for a while now and have been able to converse quite fluently.”
Comment from 9 Jul 2026 in “GPT‑Live”
What people warn about
Each mark is one comment
Loses context between messages
“I can already have impromptu practice sessions with both in Mandarin and English but they quickly lose the plot regarding their role as a tutor”
Comment from 24 Dec 2025 in “Spaced Repetition for Efficient Learning” “It's the only model that sometimes loses/forgets context in literally next message.”
Comment from 16 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking” Misidentifies or mistranslates languages
“in almost every test I did with Gemini 3.5 Live, it classified my Swedish (native) as Dutch”
Comment from 28 Aug 2026 in “Gemini-3.5-Transcribe” “Gemini responds in what I think is Spanish, or perhaps Portuguese.”
Comment from 11 Dec 2025 in “GPT-5.2” “translation technology told to guess what's most likely next instead of translate what it was given”
Comment from 15 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking” Confused by background speakers
“it wasn't as useful as you'd think. I would be talking to someone, and in the background someone else would be talking”
Comment from 15 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking” “It should be able to distinguish nearby people from people farther away and give me options to set a threshold on who to include.”
Comment from 15 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking” Hard API integration and latency
“Trying to implement their gRPC api from their specs and protobufs for Live is an exercise in immense frustration and futility”
Comment from 11 Dec 2025 in “Getting a Gemini API key is an exercise in frustration” “Sadly it doesn't support SIP. I found forwarding streams with websocket for phone calls tend to introduce some unwanted latency”
Comment from 16 Sep 2026 in “Gemini 3.8 Live and 3.8 Live Extended Thinking”
Other voice AI models people recommend
The four most recommended other voice AI models, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| GPT-Live | 72 | 48 | 60% | 4 | — | — |
| Gemini LiveThis page | 50 | 23 | 68% | 9 | — | Approved |
| Nvidia PersonaPlex 7B | 14 | 8 | 64% | 4 | — | — |
| gpt-realtime | 6 | 6 | 50% | 4 | — | — |
| Sesame voice demo | 5 | 1 | 83% | 3 | — | — |
What people said
40 of the 78 comments that mention it, newest first. Each links to the original comment.
“its working fantasticly, switching between german/english with no effort”
“Gemini Live now can be used to talk to any article on the web.”
“Gemini was already great at languages and just general spoken conversations.”
“I was incredibly surprised, its shona was really formal but that makes sense”
“Sadly it doesn't support SIP. I found forwarding streams with websocket for phone calls tend to introduce some unwanted latency”
“It does support function calling, so you can hook it to whatever tools you want.”
“what impressed me was its ability to recognize the subtle dialects of shona being spoken”
“Benchmarks don't capture the magic of having a 24/7 native-speaking conversational partner in a rare language.”
“Tell Gemini Live which speaker you want to engage with, and it is relatively intelligent about it.”
“I tried asking the model multiple internet humor and memes and it answered correctly around 80% of the time”
“It's the only model that sometimes loses/forgets context in literally next message.”
“It feels really smooth to drive, kind of buttery compared to other frontier experiences”
“Its 200-300 TPS. GPT is at 50-60. And its like 5% worse? Yeah that is a good tradeoff.”
“translation technology told to guess what's most likely next instead of translate what it was given”
“3.8 has been good enough that I don't think I'll be extending my Claude subscription”
“It should be able to distinguish nearby people from people farther away and give me options to set a threshold on who to include.”
“Playing a legal game of chess without using a guided decoding technique is a massive achievement.”
“3.8 is an incredible model even better is the infrs they host for it.”
“it wasn't as useful as you'd think. I would be talking to someone, and in the background someone else would be talking”
“Also appears to do well in other languages”
“Copes well with thick accent, voices are pleasant and latency seems low.”
“A regular search engine would have taken me hours to find an answer.”
“It's also the only model that generates accurate translation and localization.”
“Their live models have been and continue to be at the frontier. I like them a lot!”
“It really does feel like talking to a real person.”
“when you interrupt it, it shuts up immediately”
“The latency isn't too bad, but it's definitely there.”
“in almost every test I did with Gemini 3.5 Live, it classified my Swedish (native) as Dutch”
“I've been using Gemini Live for a while now and have been able to converse quite fluently.”
“it helped me verify some details around which nvme slot to use first”
“Gemini live is also pretty decent at computer vision stuff too. I've used it while working on my bike & car a few times”
“it really works surprisingly well, especially the interruption”
“I am so thrilled to see it actually helping people in the wild!”
“the live audio/video back and forth we had with the AI was almost magical”
“It kinda was fun, like a very patient professor stand right besides you. It was the one of the best math learning experience I've ever had”
“I can already have impromptu practice sessions with both in Mandarin and English but they quickly lose the plot regarding their role as a tutor”
“I was also surprised at how good Gemini Live mode is in regional Indian languages”
“The voices have none of the annoying artificial uptalking intonations that ChatGPT has”
“Gemini responds in what I think is Spanish, or perhaps Portuguese.”
“Trying to implement their gRPC api from their specs and protobufs for Live is an exercise in immense frustration and futility”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.