# Microsoft VibeVoice

10 Hacker News commenters recommended Microsoft VibeVoice and 16 criticized it, November 2025 to November 2026. It was recommended in 4 of 13 months.

**Verdict:** Free, MIT-licensed Microsoft speech model praised for expressive output and built-in diarization; critics report random music artifacts and thin documentation.

## Why people recommend it

- Expressive, good-sounding speech (5 comments): "It has some perks, is a bit more expressive in some cases, but overall is trained on really noisy data" ([source](https://news.ycombinator.com/item?id=47937500)) "In my experience, VibeVoice is quite good as well. Even the smaller model." ([source](https://news.ycombinator.com/item?id=45881777)) "it's really great frontier open-source voice AI models" ([source](https://news.ycombinator.com/item?id=47949381))
- Built-in diarization and long transcription (2 comments): "the built-in diarization is the one thing that actually caught my attention here" ([source](https://news.ycombinator.com/item?id=47942770)) "The 60-minute single-pass transcription is the part that actually matters." ([source](https://news.ycombinator.com/item?id=47939144))
- Open-source with MIT license (2 comments): "it's really great frontier open-source voice AI models" ([source](https://news.ycombinator.com/item?id=47949381)) "At least it's MIT licensed!" ([source](https://news.ycombinator.com/item?id=47935007))

## What people warn about

- Noisy training data, random music (3 comments): "it seems this was trained on large amount of podcasts with ad jingles or phone call queues with elevator music" ([source](https://news.ycombinator.com/item?id=47939255)) "It has some perks, is a bit more expressive in some cases, but overall is trained on really noisy data" ([source](https://news.ycombinator.com/item?id=47937500)) "First of all 1.5B model documentation is not there." ([source](https://news.ycombinator.com/item?id=47935668))
- Inconsistent or poor output quality (3 comments): "the quantized version often goes off the rails and loses consistency in voice tone or accent" ([source](https://news.ycombinator.com/item?id=45881648)) "a slow, unoriginal product that doesn't actually do what it advertises" ([source](https://news.ycombinator.com/item?id=47937212)) "giving me very poor results" ([source](https://news.ycombinator.com/item?id=47934165))
- Missing documentation and frontend (2 comments): "It would have been better if they provided not just weights, but also some frontend where it is usable as is." ([source](https://news.ycombinator.com/item?id=47935994)) "First of all 1.5B model documentation is not there." ([source](https://news.ycombinator.com/item?id=47935668))
- Openness claims; 7B model pulled (2 comments): "MS is misrepresenting their openness here" ([source](https://news.ycombinator.com/item?id=47936001)) "they rug-pulled the 7B for our own safety" ([source](https://news.ycombinator.com/item?id=47936293))

## Alternatives hackers recommend

- [ElevenLabs](https://hackerapproved.com/p/elevenlabs.md): 59 recommend
- [Kokoro TTS](https://hackerapproved.com/p/kokoro-tts.md): 59 recommend
- [Pocket TTS](https://hackerapproved.com/p/pocket-tts.md): 41 recommend
- [Qwen3-TTS](https://hackerapproved.com/p/qwen3-tts.md): 41 recommend
- [Kitten TTS](https://hackerapproved.com/p/kitten-tts.md): 33 recommend

## Cite as

Hacker Approved, "Microsoft VibeVoice", https://hackerapproved.com/p/vibevoice/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).