{"slug": "kokoro", "name": "Kokoro", "category": "ai", "category_name": "AI models and assistants", "description": "Kokoro-82M, a small open-weight text-to-speech model that runs locally.", "recommenders": 15, "commenters": 23, "warners": 0, "score": 16.03, "rising": -0.93, "first": "2025-12-26", "latest": "2026-10-25", "why": "Liked as natural-sounding local text-to-speech that is small enough for consumer hardware and phones, often used to turn ebooks and articles into audio.", "why_is_quote": false, "has_verdict": true, "sentiment": {"recommend": 15, "neutral": 7, "compare": 1, "warn": 0}, "spark": [0, 1, 0, 1, 0, 0, 0, 0, 10, 0, 1, 2], "listed": true, "loved": ["Natural voice quality", "Audiobooks and articles"], "caveat": "Pronunciation errors", "short_quote": "I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent", "icon": true, "domain": null, "github_repo": "github.com/nazdridoy/kokoro-tts", "official_url": "https://huggingface.co/hexgrad/Kokoro-82M", "pricing": "open source", "aliases": ["Kokoro", "Kokoro TTS", "Kokoro-82M", "Kokoro 82M"], "verdict": [{"text": "Liked as natural-sounding local text-to-speech", "cites": [{"id": 802719744, "item_id": 50169984, "url": "https://news.ycombinator.com/item?id=50169984", "by": "gavinray", "date": "2026-10-20", "stance": "recommend", "quote": "Best sounding local TTS engine I've heard by far is Kokoro", "excerpt": "Best sounding local TTS engine I've heard by far is Kokoro It's small enough to be run on modern phones realtime, I use it for listening to PDF's/EPUB as Audiobooks https://huggingface.co/hexgrad/Kokoro-82M", "context": "", "use_case": "listening to PDFs/EPUBs as audiobooks", "switched_from": "", "story": "Voxtral 3"}, {"id": 781191584, "item_id": 48824474, "url": "https://news.ycombinator.com/item?id=48824474", "by": "zackify", "date": "2026-07-07", "stance": "recommend", "quote": "I use kokoro with home assistant and its great. I find its the most natural sounding and small too.", "excerpt": "I use kokoro with home assistant and its great. I find its the most natural sounding and small too. I speak over sonos speakers when certain events happen. And use it as my voice assistant.", "context": "", "use_case": "voice assistant TTS", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781203648, "item_id": 48825228, "url": "https://news.ycombinator.com/item?id=48825228", "by": "rnxrx", "date": "2026-07-07", "stance": "recommend", "quote": "I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent", "excerpt": "Another endorsement - I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent, both on- and off- GPU. Even with Elevenlabs (long time subscriber) the comparative quality of Kokoro keeps up really well until you get to their larger models with their professional voices. I do wish there were better support for SSML, as well as deeper documentation of how to influence inf", "context": "", "use_case": "TTS in an app", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 751149184, "item_id": 46946824, "url": "https://news.ycombinator.com/item?id=46946824", "by": "jordanf", "date": "2026-02-09", "stance": "recommend", "quote": "it's actually unbelievably good", "excerpt": "thank you! it's actually unbelievably good. check out the demo audio files here: https://github.com/nazdridoy/kokoro-tts?tab=readme-ov-file#demo", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Ask HN: What Are You Working On? (February 2026)"}]}, {"text": " that is small enough for consumer hardware and phones,", "cites": [{"id": 802719744, "item_id": 50169984, "url": "https://news.ycombinator.com/item?id=50169984", "by": "gavinray", "date": "2026-10-20", "stance": "recommend", "quote": "Best sounding local TTS engine I've heard by far is Kokoro", "excerpt": "Best sounding local TTS engine I've heard by far is Kokoro It's small enough to be run on modern phones realtime, I use it for listening to PDF's/EPUB as Audiobooks https://huggingface.co/hexgrad/Kokoro-82M", "context": "", "use_case": "listening to PDFs/EPUBs as audiobooks", "switched_from": "", "story": "Voxtral 3"}, {"id": 781191584, "item_id": 48824474, "url": "https://news.ycombinator.com/item?id=48824474", "by": "zackify", "date": "2026-07-07", "stance": "recommend", "quote": "I use kokoro with home assistant and its great. I find its the most natural sounding and small too.", "excerpt": "I use kokoro with home assistant and its great. I find its the most natural sounding and small too. I speak over sonos speakers when certain events happen. And use it as my voice assistant.", "context": "", "use_case": "voice assistant TTS", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781178880, "item_id": 48823680, "url": "https://news.ycombinator.com/item?id=48823680", "by": "Judson", "date": "2026-07-07", "stance": "recommend", "quote": "Love Kokoro tts.", "excerpt": "Love Kokoro tts. I wrote https://github.com/Jud/kokoro-coreml to try pushing the limits a bit on speed & size. Such great quality at a given size. As others have mentioned short utterances are problematic, but solvable.", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}]}, {"text": " often used to turn ebooks and articles into audio.", "cites": [{"id": 802719744, "item_id": 50169984, "url": "https://news.ycombinator.com/item?id=50169984", "by": "gavinray", "date": "2026-10-20", "stance": "recommend", "quote": "Best sounding local TTS engine I've heard by far is Kokoro", "excerpt": "Best sounding local TTS engine I've heard by far is Kokoro It's small enough to be run on modern phones realtime, I use it for listening to PDF's/EPUB as Audiobooks https://huggingface.co/hexgrad/Kokoro-82M", "context": "", "use_case": "listening to PDFs/EPUBs as audiobooks", "switched_from": "", "story": "Voxtral 3"}, {"id": 781267264, "item_id": 48829204, "url": "https://news.ycombinator.com/item?id=48829204", "by": "lucumo", "date": "2026-07-08", "stance": "recommend", "quote": "I love Kokoro. I use it to read ebooks that don't have audiobooks", "excerpt": "I love Kokoro. I use it to read ebooks that don't have audiobooks and it works pretty good for that. I have a Python script that reads epub/html, runs the model and writes mp3s. Kokoro supports a limited number of languages. I've had to resort to other models to support Dutch, and they just aren't as good. And usually a lot slower and bigger (too big for my 8GB VRAM). I've had to resort to the Windows built-in TTS. I", "context": "", "use_case": "reading ebooks aloud", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 797142800, "item_id": 49821425, "url": "https://news.ycombinator.com/item?id=49821425", "by": "exhilaration", "date": "2026-09-23", "stance": "recommend", "quote": "Personally I use Kokoro with a python front-end on my Macbook", "excerpt": "Browser extension? Try these: http://www.paper2audio.com/ https://freevoicereader.com/ https://chromewebstore.google.com/detail/sza-text-to-speech/jfaocnefnpgpdmfodmpcmlmfllicbpbi?pli=1 https://chromewebstore.google.com/detail/flow-tts/hknidankihadgkfhohimfmfiamdeiedb I found all these on this subreddit: https://www.reddit.com/r/TextToSpeech/ I've also see comments like \"Microsoft Edge's read-aloud feature is amazing", "context": "listens to articles as audiobooks", "use_case": "text-to-speech to MP3 audiobooks", "switched_from": "", "story": "Gemini 3.8 text-to-speech"}, {"id": 775375360, "item_id": 48460960, "url": "https://news.ycombinator.com/item?id=48460960", "by": "littlexsparkee", "date": "2026-06-09", "stance": "neutral", "quote": "I do something similar with Kokoro, Cloudscraper, and Trafilatura", "excerpt": "I do something similar with Kokoro, Cloudscraper, and Trafilatura but just share the output to my phone with Airdrop. I use it for articles, PDFs, and HN threads.", "context": "", "use_case": "text-to-speech for articles", "switched_from": "", "story": "Ask HN: What are tools you have made for yourself since the advent of AI"}]}, {"text": " The most common caveats are mistakes on some phrases and short utterances,", "cites": [{"id": 803737376, "item_id": 50233586, "url": "https://news.ycombinator.com/item?id=50233586", "by": "tsycho", "date": "2026-10-25", "stance": "recommend", "quote": "I use Kokoro 82M and it's good overall, but it does make mistakes on some common phrases", "excerpt": "What STT and TTS models do you recommend that would run well on consumer hardware? I use Kokoro 82M and it's good overall, but it does make mistakes on some common phrases, and it's not real time fast (runs at about 6-7X content-to-encoding time ratio on my home server)", "context": "runs a home server", "use_case": "text-to-speech on home server", "switched_from": "", "story": "Dessa: Conversation Model trained only on speech"}, {"id": 781178880, "item_id": 48823680, "url": "https://news.ycombinator.com/item?id=48823680", "by": "Judson", "date": "2026-07-07", "stance": "recommend", "quote": "Love Kokoro tts.", "excerpt": "Love Kokoro tts. I wrote https://github.com/Jud/kokoro-coreml to try pushing the limits a bit on speed & size. Such great quality at a given size. As others have mentioned short utterances are problematic, but solvable.", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}]}, {"text": " plus limited language support.", "cites": [{"id": 781267264, "item_id": 48829204, "url": "https://news.ycombinator.com/item?id=48829204", "by": "lucumo", "date": "2026-07-08", "stance": "recommend", "quote": "I love Kokoro. I use it to read ebooks that don't have audiobooks", "excerpt": "I love Kokoro. I use it to read ebooks that don't have audiobooks and it works pretty good for that. I have a Python script that reads epub/html, runs the model and writes mp3s. Kokoro supports a limited number of languages. I've had to resort to other models to support Dutch, and they just aren't as good. And usually a lot slower and bigger (too big for my 8GB VRAM). I've had to resort to the Windows built-in TTS. I", "context": "", "use_case": "reading ebooks aloud", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}]}], "praise": [{"label": "Natural voice quality", "summary": "Commenters call it the best-sounding or most natural local TTS, holding up well against paid services.", "quotes": [{"id": 802719744, "item_id": 50169984, "url": "https://news.ycombinator.com/item?id=50169984", "by": "gavinray", "date": "2026-10-20", "stance": "recommend", "quote": "Best sounding local TTS engine I've heard by far is Kokoro", "excerpt": "Best sounding local TTS engine I've heard by far is Kokoro It's small enough to be run on modern phones realtime, I use it for listening to PDF's/EPUB as Audiobooks https://huggingface.co/hexgrad/Kokoro-82M", "context": "", "use_case": "listening to PDFs/EPUBs as audiobooks", "switched_from": "", "story": "Voxtral 3"}, {"id": 781191584, "item_id": 48824474, "url": "https://news.ycombinator.com/item?id=48824474", "by": "zackify", "date": "2026-07-07", "stance": "recommend", "quote": "I use kokoro with home assistant and its great. I find its the most natural sounding and small too.", "excerpt": "I use kokoro with home assistant and its great. I find its the most natural sounding and small too. I speak over sonos speakers when certain events happen. And use it as my voice assistant.", "context": "", "use_case": "voice assistant TTS", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781203648, "item_id": 48825228, "url": "https://news.ycombinator.com/item?id=48825228", "by": "rnxrx", "date": "2026-07-07", "stance": "recommend", "quote": "I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent", "excerpt": "Another endorsement - I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent, both on- and off- GPU. Even with Elevenlabs (long time subscriber) the comparative quality of Kokoro keeps up really well until you get to their larger models with their professional voices. I do wish there were better support for SSML, as well as deeper documentation of how to influence inf", "context": "", "use_case": "TTS in an app", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}]}, {"label": "Audiobooks and articles", "summary": "Widely used to listen to ebooks, PDFs, articles and HN threads as audio.", "quotes": [{"id": 781267264, "item_id": 48829204, "url": "https://news.ycombinator.com/item?id=48829204", "by": "lucumo", "date": "2026-07-08", "stance": "recommend", "quote": "I love Kokoro. I use it to read ebooks that don't have audiobooks", "excerpt": "I love Kokoro. I use it to read ebooks that don't have audiobooks and it works pretty good for that. I have a Python script that reads epub/html, runs the model and writes mp3s. Kokoro supports a limited number of languages. I've had to resort to other models to support Dutch, and they just aren't as good. And usually a lot slower and bigger (too big for my 8GB VRAM). I've had to resort to the Windows built-in TTS. I", "context": "", "use_case": "reading ebooks aloud", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 802719744, "item_id": 50169984, "url": "https://news.ycombinator.com/item?id=50169984", "by": "gavinray", "date": "2026-10-20", "stance": "recommend", "quote": "Best sounding local TTS engine I've heard by far is Kokoro", "excerpt": "Best sounding local TTS engine I've heard by far is Kokoro It's small enough to be run on modern phones realtime, I use it for listening to PDF's/EPUB as Audiobooks https://huggingface.co/hexgrad/Kokoro-82M", "context": "", "use_case": "listening to PDFs/EPUBs as audiobooks", "switched_from": "", "story": "Voxtral 3"}, {"id": 775375360, "item_id": 48460960, "url": "https://news.ycombinator.com/item?id=48460960", "by": "littlexsparkee", "date": "2026-06-09", "stance": "neutral", "quote": "I do something similar with Kokoro, Cloudscraper, and Trafilatura", "excerpt": "I do something similar with Kokoro, Cloudscraper, and Trafilatura but just share the output to my phone with Airdrop. I use it for articles, PDFs, and HN threads.", "context": "", "use_case": "text-to-speech for articles", "switched_from": "", "story": "Ask HN: What are tools you have made for yourself since the advent of AI"}]}, {"label": "Small and local", "summary": "Small enough to run on phones, laptops and home servers, with or without a GPU.", "quotes": [{"id": 802719744, "item_id": 50169984, "url": "https://news.ycombinator.com/item?id=50169984", "by": "gavinray", "date": "2026-10-20", "stance": "recommend", "quote": "Best sounding local TTS engine I've heard by far is Kokoro", "excerpt": "Best sounding local TTS engine I've heard by far is Kokoro It's small enough to be run on modern phones realtime, I use it for listening to PDF's/EPUB as Audiobooks https://huggingface.co/hexgrad/Kokoro-82M", "context": "", "use_case": "listening to PDFs/EPUBs as audiobooks", "switched_from": "", "story": "Voxtral 3"}, {"id": 781203648, "item_id": 48825228, "url": "https://news.ycombinator.com/item?id=48825228", "by": "rnxrx", "date": "2026-07-07", "stance": "recommend", "quote": "I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent", "excerpt": "Another endorsement - I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent, both on- and off- GPU. Even with Elevenlabs (long time subscriber) the comparative quality of Kokoro keeps up really well until you get to their larger models with their professional voices. I do wish there were better support for SSML, as well as deeper documentation of how to influence inf", "context": "", "use_case": "TTS in an app", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781186608, "item_id": 48824163, "url": "https://news.ycombinator.com/item?id=48824163", "by": "janpmz", "date": "2026-07-07", "stance": "recommend", "quote": "Kokoro for TTS", "excerpt": "Both Text-to-Speech and Speech-to-Text now have local models that are good enough to get the job done. Kokoro for TTS, Parakeet for STT and Fluid-1 for text formatting (I use it with FluidVoice). I hope this is a trend that continues for other applications.", "context": "", "use_case": "local text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}]}, {"label": "Easy to build on", "summary": "Developers have wrapped it in MCP servers, CoreML ports, CLIs and APIs, and tuned it to run faster.", "quotes": [{"id": 781203328, "item_id": 48825208, "url": "https://news.ycombinator.com/item?id=48825208", "by": "mberg", "date": "2026-07-07", "stance": "recommend", "quote": "Kokoro is great.", "excerpt": "Kokoro is great. I built an mcp for it a while back that has gotten decent traction - https://github.com/mberg/kokoro-tts-mcp if anyone wants to go that route", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781178880, "item_id": 48823680, "url": "https://news.ycombinator.com/item?id=48823680", "by": "Judson", "date": "2026-07-07", "stance": "recommend", "quote": "Love Kokoro tts.", "excerpt": "Love Kokoro tts. I wrote https://github.com/Jud/kokoro-coreml to try pushing the limits a bit on speed & size. Such great quality at a given size. As others have mentioned short utterances are problematic, but solvable.", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781183152, "item_id": 48823947, "url": "https://news.ycombinator.com/item?id=48823947", "by": "deivid", "date": "2026-07-07", "stance": "recommend", "quote": "it runs 3x faster.\n\nQuality is very close.", "excerpt": "I spent a day fiddling with AI and dropping the expensive layers in kokoro, on phones, on CPU, on MNN, it runs 3x faster. Quality is very close. Will vary in your setup, but here is my script: https://github.com/DavidVentura/translator-rs/blob/master/scripts/kokoro_mnn/rebuild_best_mnn.sh#L53", "context": "", "use_case": "on-device text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}]}, {"label": "Home voice assistants", "summary": "Used as the TTS for Home Assistant voice setups.", "quotes": [{"id": 781191584, "item_id": 48824474, "url": "https://news.ycombinator.com/item?id=48824474", "by": "zackify", "date": "2026-07-07", "stance": "recommend", "quote": "I use kokoro with home assistant and its great. I find its the most natural sounding and small too.", "excerpt": "I use kokoro with home assistant and its great. I find its the most natural sounding and small too. I speak over sonos speakers when certain events happen. And use it as my voice assistant.", "context": "", "use_case": "voice assistant TTS", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 800362642, "item_id": 50022665, "url": "https://news.ycombinator.com/item?id=50022665", "by": "cco", "date": "2026-10-09", "stance": "neutral", "quote": "Kokoro for TTS", "excerpt": "Yeah maybe I'll play around with different precision, I did see improvement when I went from fp4 to full fidelity though. This is my setup for Home Assistant voice, parakeet for STT, GLM 5.3 to generate responses, and Kokoro for TTS. I use local routing for simple commands like turn on the lights, but if I want to ask questions or something, it gets routed to GLM 5.3. It is basically free, half a cent per turn maybe,", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Whistle: Speech to Text in 16.9 MB"}]}], "caveats": [{"label": "Pronunciation errors", "summary": "It makes mistakes on some common phrases, and short utterances are problematic.", "quotes": [{"id": 803737376, "item_id": 50233586, "url": "https://news.ycombinator.com/item?id=50233586", "by": "tsycho", "date": "2026-10-25", "stance": "recommend", "quote": "I use Kokoro 82M and it's good overall, but it does make mistakes on some common phrases", "excerpt": "What STT and TTS models do you recommend that would run well on consumer hardware? I use Kokoro 82M and it's good overall, but it does make mistakes on some common phrases, and it's not real time fast (runs at about 6-7X content-to-encoding time ratio on my home server)", "context": "runs a home server", "use_case": "text-to-speech on home server", "switched_from": "", "story": "Dessa: Conversation Model trained only on speech"}, {"id": 781178880, "item_id": 48823680, "url": "https://news.ycombinator.com/item?id=48823680", "by": "Judson", "date": "2026-07-07", "stance": "recommend", "quote": "Love Kokoro tts.", "excerpt": "Love Kokoro tts. I wrote https://github.com/Jud/kokoro-coreml to try pushing the limits a bit on speed & size. Such great quality at a given size. As others have mentioned short utterances are problematic, but solvable.", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}]}, {"label": "Limited languages", "summary": "It supports a limited number of languages, so Dutch, for example, needed other models.", "quotes": [{"id": 781267264, "item_id": 48829204, "url": "https://news.ycombinator.com/item?id=48829204", "by": "lucumo", "date": "2026-07-08", "stance": "recommend", "quote": "I love Kokoro. I use it to read ebooks that don't have audiobooks", "excerpt": "I love Kokoro. I use it to read ebooks that don't have audiobooks and it works pretty good for that. I have a Python script that reads epub/html, runs the model and writes mp3s. Kokoro supports a limited number of languages. I've had to resort to other models to support Dutch, and they just aren't as good. And usually a lot slower and bigger (too big for my 8GB VRAM). I've had to resort to the Windows built-in TTS. I", "context": "", "use_case": "reading ebooks aloud", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}]}, {"label": "Limited control", "summary": "One user wants better SSML support and documentation on controlling inflection.", "quotes": [{"id": 781203648, "item_id": 48825228, "url": "https://news.ycombinator.com/item?id=48825228", "by": "rnxrx", "date": "2026-07-07", "stance": "recommend", "quote": "I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent", "excerpt": "Another endorsement - I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent, both on- and off- GPU. Even with Elevenlabs (long time subscriber) the comparative quality of Kokoro keeps up really well until you get to their larger models with their professional voices. I do wish there were better support for SSML, as well as deeper documentation of how to influence inf", "context": "", "use_case": "TTS in an app", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}]}, {"label": "Speed varies", "summary": "One home-server user says it is not real-time fast.", "quotes": [{"id": 803737376, "item_id": 50233586, "url": "https://news.ycombinator.com/item?id=50233586", "by": "tsycho", "date": "2026-10-25", "stance": "recommend", "quote": "I use Kokoro 82M and it's good overall, but it does make mistakes on some common phrases", "excerpt": "What STT and TTS models do you recommend that would run well on consumer hardware? I use Kokoro 82M and it's good overall, but it does make mistakes on some common phrases, and it's not real time fast (runs at about 6-7X content-to-encoding time ratio on my home server)", "context": "runs a home server", "use_case": "text-to-speech on home server", "switched_from": "", "story": "Dessa: Conversation Model trained only on speech"}]}, {"label": "Alternatives", "summary": "One commenter finds pocket-tts much better.", "quotes": [{"id": 781157264, "item_id": 48822329, "url": "https://news.ycombinator.com/item?id=48822329", "by": "teravor", "date": "2026-07-07", "stance": "compare", "quote": "kokoro is decent but pocket-tts is much better", "excerpt": "kokoro is decent but pocket-tts is much better especially when you rip a good voice. https://github.com/kyutai-labs/pocket-tts the onnx version of pocket-tts does perform better. https://huggingface.co/KevinAHM/pocket-tts-onnx", "context": "", "use_case": "local text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}]}], "who": [{"by": "tsycho", "context": "runs a home server", "url": "https://news.ycombinator.com/item?id=50233586", "date": "2026-10-25"}, {"by": "exhilaration", "context": "listens to articles as audiobooks", "url": "https://news.ycombinator.com/item?id=49821425", "date": "2026-09-23"}, {"by": "0gs", "context": "Runs an LLM comedy podcast", "url": "https://news.ycombinator.com/item?id=48822554", "date": "2026-07-07"}, {"by": "cbrews", "context": "audio background", "url": "https://news.ycombinator.com/item?id=46739544", "date": "2026-01-23"}], "tags": ["for audiobooks", "runs locally", "for Home Assistant", "for Mac users", "for app developers"], "use_cases": ["text-to-speech", "local text-to-speech", "text-to-speech on home server", "listening to pdfs/epubs as audiobooks", "text-to-speech model", "text-to-speech to mp3 audiobooks"], "compared": [{"slug": "home-assistant", "name": "Home Assistant", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "parakeet", "name": "Parakeet", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "whisper", "name": "Whisper", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "gemma", "name": "Gemma", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=48821576", "title": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro", "mentions": 11}, {"url": "https://news.ycombinator.com/item?id=46731068", "title": "Ask HN: What's the current best local/open speech-to-speech setup?", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=50213925", "title": "Dessa: Conversation Model trained only on speech", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=50168396", "title": "Voxtral 3", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=50069337", "title": "Show HN: Use your OpenAI subscription to create audiobooks", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=50008427", "title": "Whistle: Speech to Text in 16.9 MB", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {"recommend": 1}, "2026-01": {"neutral": 2}, "2026-02": {"recommend": 1}, "2026-03": {}, "2026-04": {}, "2026-05": {"neutral": 1}, "2026-06": {"neutral": 1}, "2026-07": {"compare": 1, "recommend": 10, "neutral": 1}, "2026-08": {}, "2026-09": {"recommend": 1}, "2026-10": {"neutral": 2, "recommend": 2}}, "quotes": [{"id": 803737376, "item_id": 50233586, "url": "https://news.ycombinator.com/item?id=50233586", "by": "tsycho", "date": "2026-10-25", "stance": "recommend", "quote": "I use Kokoro 82M and it's good overall, but it does make mistakes on some common phrases", "excerpt": "What STT and TTS models do you recommend that would run well on consumer hardware? I use Kokoro 82M and it's good overall, but it does make mistakes on some common phrases, and it's not real time fast (runs at about 6-7X content-to-encoding time ratio on my home server)", "context": "runs a home server", "use_case": "text-to-speech on home server", "switched_from": "", "story": "Dessa: Conversation Model trained only on speech"}, {"id": 802719744, "item_id": 50169984, "url": "https://news.ycombinator.com/item?id=50169984", "by": "gavinray", "date": "2026-10-20", "stance": "recommend", "quote": "Best sounding local TTS engine I've heard by far is Kokoro", "excerpt": "Best sounding local TTS engine I've heard by far is Kokoro It's small enough to be run on modern phones realtime, I use it for listening to PDF's/EPUB as Audiobooks https://huggingface.co/hexgrad/Kokoro-82M", "context": "", "use_case": "listening to PDFs/EPUBs as audiobooks", "switched_from": "", "story": "Voxtral 3"}, {"id": 801284241, "item_id": 50080265, "url": "https://news.ycombinator.com/item?id=50080265", "by": "mier", "date": "2026-10-13", "stance": "neutral", "quote": "using MLX and Kokoro", "excerpt": "I use my hobby app PageHum to read short to medium articles in devices using MLX and Kokoro instead of. Give it a try if anyone interested in https://github.com/Mieraidihaimu/pagehum (Xcode 27+)", "context": "", "use_case": "text-to-speech model", "switched_from": "", "story": "Show HN: Use your OpenAI subscription to create audiobooks"}, {"id": 800362642, "item_id": 50022665, "url": "https://news.ycombinator.com/item?id=50022665", "by": "cco", "date": "2026-10-09", "stance": "neutral", "quote": "Kokoro for TTS", "excerpt": "Yeah maybe I'll play around with different precision, I did see improvement when I went from fp4 to full fidelity though. This is my setup for Home Assistant voice, parakeet for STT, GLM 5.3 to generate responses, and Kokoro for TTS. I use local routing for simple commands like turn on the lights, but if I want to ask questions or something, it gets routed to GLM 5.3. It is basically free, half a cent per turn maybe,", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Whistle: Speech to Text in 16.9 MB"}, {"id": 797142800, "item_id": 49821425, "url": "https://news.ycombinator.com/item?id=49821425", "by": "exhilaration", "date": "2026-09-23", "stance": "recommend", "quote": "Personally I use Kokoro with a python front-end on my Macbook", "excerpt": "Browser extension? Try these: http://www.paper2audio.com/ https://freevoicereader.com/ https://chromewebstore.google.com/detail/sza-text-to-speech/jfaocnefnpgpdmfodmpcmlmfllicbpbi?pli=1 https://chromewebstore.google.com/detail/flow-tts/hknidankihadgkfhohimfmfiamdeiedb I found all these on this subreddit: https://www.reddit.com/r/TextToSpeech/ I've also see comments like \"Microsoft Edge's read-aloud feature is amazing", "context": "listens to articles as audiobooks", "use_case": "text-to-speech to MP3 audiobooks", "switched_from": "", "story": "Gemini 3.8 text-to-speech"}, {"id": 781606465, "item_id": 48850404, "url": "https://news.ycombinator.com/item?id=48850404", "by": "rpac0", "date": "2026-07-09", "stance": "neutral", "quote": "also uses Kokoro-82M", "excerpt": "I wrote a CLI-based Text-to-Speech that also uses Kokoro-82M. I've been using this everyday and HN folks might find this useful too. https://github.com/realpacific/readthis", "context": "", "use_case": "text-to-speech engine", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781303715, "item_id": 48831482, "url": "https://news.ycombinator.com/item?id=48831482", "by": "nekusar", "date": "2026-07-08", "stance": "recommend", "quote": "For my STT and TTS, ones on the HN front page right now", "excerpt": "Thank you for addressing my list. I'd like to use it, but requirements on non-local services is a deal breaker for me. Right now, with my Nanoclaw, I'm running it in a container on a VM on proxmox. If things get bad, I lose a backed up machine and any services it has credentials to. I can understand why IMAP is a pain. There's plenty of functions that can trash mailboxes. I don't envy you all in this task. For my STT", "context": "", "use_case": "local text-to-speech", "switched_from": "", "story": "Show HN: Rowboat \u2013 Open-source, local-first alternative to Claude Desktop"}, {"id": 781267264, "item_id": 48829204, "url": "https://news.ycombinator.com/item?id=48829204", "by": "lucumo", "date": "2026-07-08", "stance": "recommend", "quote": "I love Kokoro. I use it to read ebooks that don't have audiobooks", "excerpt": "I love Kokoro. I use it to read ebooks that don't have audiobooks and it works pretty good for that. I have a Python script that reads epub/html, runs the model and writes mp3s. Kokoro supports a limited number of languages. I've had to resort to other models to support Dutch, and they just aren't as good. And usually a lot slower and bigger (too big for my 8GB VRAM). I've had to resort to the Windows built-in TTS. I", "context": "", "use_case": "reading ebooks aloud", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781205872, "item_id": 48825367, "url": "https://news.ycombinator.com/item?id=48825367", "by": "karimf", "date": "2026-07-07", "stance": "recommend", "quote": "Kokoro is a really good model, considered it\u2019s released 1.5 years ago. It\u2019s punching above its weight", "excerpt": "This repo is a good starting point for comparing TTS models https://github.com/5uck1ess/tts-bench Kokoro is a really good model, considered it\u2019s released 1.5 years ago. It\u2019s punching above its weight https://5uck1ess.github.io/tts-bench/scores.html", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781203648, "item_id": 48825228, "url": "https://news.ycombinator.com/item?id=48825228", "by": "rnxrx", "date": "2026-07-07", "stance": "recommend", "quote": "I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent", "excerpt": "Another endorsement - I used Kokoro pretty extensively with an app I was developing over the last year and it's been excellent, both on- and off- GPU. Even with Elevenlabs (long time subscriber) the comparative quality of Kokoro keeps up really well until you get to their larger models with their professional voices. I do wish there were better support for SSML, as well as deeper documentation of how to influence inf", "context": "", "use_case": "TTS in an app", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781203328, "item_id": 48825208, "url": "https://news.ycombinator.com/item?id=48825208", "by": "mberg", "date": "2026-07-07", "stance": "recommend", "quote": "Kokoro is great.", "excerpt": "Kokoro is great. I built an mcp for it a while back that has gotten decent traction - https://github.com/mberg/kokoro-tts-mcp if anyone wants to go that route", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781191584, "item_id": 48824474, "url": "https://news.ycombinator.com/item?id=48824474", "by": "zackify", "date": "2026-07-07", "stance": "recommend", "quote": "I use kokoro with home assistant and its great. I find its the most natural sounding and small too.", "excerpt": "I use kokoro with home assistant and its great. I find its the most natural sounding and small too. I speak over sonos speakers when certain events happen. And use it as my voice assistant.", "context": "", "use_case": "voice assistant TTS", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781186608, "item_id": 48824163, "url": "https://news.ycombinator.com/item?id=48824163", "by": "janpmz", "date": "2026-07-07", "stance": "recommend", "quote": "Kokoro for TTS", "excerpt": "Both Text-to-Speech and Speech-to-Text now have local models that are good enough to get the job done. Kokoro for TTS, Parakeet for STT and Fluid-1 for text formatting (I use it with FluidVoice). I hope this is a trend that continues for other applications.", "context": "", "use_case": "local text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781183152, "item_id": 48823947, "url": "https://news.ycombinator.com/item?id=48823947", "by": "deivid", "date": "2026-07-07", "stance": "recommend", "quote": "it runs 3x faster.\n\nQuality is very close.", "excerpt": "I spent a day fiddling with AI and dropping the expensive layers in kokoro, on phones, on CPU, on MNN, it runs 3x faster. Quality is very close. Will vary in your setup, but here is my script: https://github.com/DavidVentura/translator-rs/blob/master/scripts/kokoro_mnn/rebuild_best_mnn.sh#L53", "context": "", "use_case": "on-device text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781178880, "item_id": 48823680, "url": "https://news.ycombinator.com/item?id=48823680", "by": "Judson", "date": "2026-07-07", "stance": "recommend", "quote": "Love Kokoro tts.", "excerpt": "Love Kokoro tts. I wrote https://github.com/Jud/kokoro-coreml to try pushing the limits a bit on speed & size. Such great quality at a given size. As others have mentioned short utterances are problematic, but solvable.", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781160864, "item_id": 48822554, "url": "https://news.ycombinator.com/item?id=48822554", "by": "0gs", "date": "2026-07-07", "stance": "recommend", "quote": "kokoro is surprisingly great at nuance", "excerpt": "kokoro is surprisingly great at nuance but it's tough to improve that last ~2% or so. kokoro + rvc is really great too; i use that for ELEMENT47, the LLM-centric comedy podcast i do that i wish more people would listen to. (e47.net , feel free to subscribe!)", "context": "Runs an LLM comedy podcast", "use_case": "podcast voice generation", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 781157264, "item_id": 48822329, "url": "https://news.ycombinator.com/item?id=48822329", "by": "teravor", "date": "2026-07-07", "stance": "compare", "quote": "kokoro is decent but pocket-tts is much better", "excerpt": "kokoro is decent but pocket-tts is much better especially when you rip a good voice. https://github.com/kyutai-labs/pocket-tts the onnx version of pocket-tts does perform better. https://huggingface.co/KevinAHM/pocket-tts-onnx", "context": "", "use_case": "local text-to-speech", "switched_from": "", "story": "Local, CPU-Friendly, High-Quality TTS (Text-to-Speech) with Kokoro"}, {"id": 775375360, "item_id": 48460960, "url": "https://news.ycombinator.com/item?id=48460960", "by": "littlexsparkee", "date": "2026-06-09", "stance": "neutral", "quote": "I do something similar with Kokoro, Cloudscraper, and Trafilatura", "excerpt": "I do something similar with Kokoro, Cloudscraper, and Trafilatura but just share the output to my phone with Airdrop. I use it for articles, PDFs, and HN threads.", "context": "", "use_case": "text-to-speech for articles", "switched_from": "", "story": "Ask HN: What are tools you have made for yourself since the advent of AI"}, {"id": 768243347, "item_id": 48015209, "url": "https://news.ycombinator.com/item?id=48015209", "by": "pncnmnp", "date": "2026-05-04", "stance": "neutral", "quote": "converts each sentence to audio via Kokoro TTS", "excerpt": "The whole setup works on my M2 MacBook Pro with 16 GB RAM. I use Gemma 4B via LiteRT-LM. I've found that LiteRT-LM has a much lower DRAM footprint than Ollama. I've also made tons of optimizations in the code - for eg, you can do quite a bit with a 16k context window for a voice assistant while managing a good footprint, so I keep track of the token usage and then perform an auto-compaction after a while. I use sub-a", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "How OpenAI delivers low-latency voice AI at scale"}, {"id": 751149184, "item_id": 46946824, "url": "https://news.ycombinator.com/item?id=46946824", "by": "jordanf", "date": "2026-02-09", "stance": "recommend", "quote": "it's actually unbelievably good", "excerpt": "thank you! it's actually unbelievably good. check out the demo audio files here: https://github.com/nazdridoy/kokoro-tts?tab=readme-ov-file#demo", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Ask HN: What Are You Working On? (February 2026)"}, {"id": 749227618, "item_id": 46826726, "url": "https://news.ycombinator.com/item?id=46826726", "by": "krunkosaurus", "date": "2026-01-30", "stance": "neutral", "quote": "Internally it uses Parakeet, Kokoro TTS, Qwen3 TTS", "excerpt": "My time to shine! I just open sourced a REST API and framework that you can point any of your projects to such as OpenClaw so you can do TTS and SST using your own hardware. It's specially focused on Mac M-class chips to utilize MLX: https://github.com/Sogni-AI/sogni-voice All free and open source. Internally it uses Parakeet, Kokoro TTS, Qwen3 TTS (with voice cloning support!) Also supports creating your own API key", "context": "", "use_case": "text to speech", "switched_from": "", "story": "Ask HN: What's the current best local/open speech-to-speech setup?"}, {"id": 747832705, "item_id": 46739544, "url": "https://news.ycombinator.com/item?id=46739544", "by": "cbrews", "date": "2026-01-23", "stance": "neutral", "quote": "Text-to-Speech (using Kokoro at the time)", "excerpt": "Thanks for sharing this! I'm going to put this on my list to play around with. I'm not really an expert in this tech, I come from the audio background, but recently was playing around with streaming Speech-to-Text (using Whisper) / Text-to-Speech (using Kokoro at the time) on a local machine. The most challenging part in my build was tuning the inference batch sizing here. I was able to get it working well for Speech", "context": "audio background", "use_case": "local text-to-speech", "switched_from": "", "story": "Ask HN: What's the current best local/open speech-to-speech setup?"}, {"id": 742366577, "item_id": 46397911, "url": "https://news.ycombinator.com/item?id=46397911", "by": "bryanstjohn", "date": "2025-12-26", "stance": "recommend", "quote": "We prioritize a self-hosted instance of Kokoro TTS (running on a Reserved VM) for high-quality, low-latency audio", "excerpt": "Hi HN, I\u2019m Bryan founder of ForwardToAudio. The Problem: Like many of you, I subscribe to way too many long-form newsletters (Substack, technical digests, etc.), but I never find the time to actually sit down and read them. I tried using standard screen readers, but listening to them read URLs, \"view in browser\" links, and footer disclaimers was unbearable. The Solution: I built a tool that creates a private podcast ", "context": "", "use_case": "text-to-speech", "switched_from": "", "story": "Show HN: ForwardToAudio \u2013 Turn newsletters into a private podcast using AI"}]}