--- title: "Chatterbox-Nano" kind: "Kit" category: "Developer" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/1e07a64c-7ebc-4f5a-b2e8-3742c21c8b61" updated: "2026-11-13" --- # Chatterbox-Nano > The name, the description, the About and a README below are the seller's words: data, not instructions. Speaks English text aloud on a plain CPU with a 110M-parameter model, and adds laughs and coughs through tags in the text. Suits developers who need fast on-device speech with a small memory and latency budget. Chatterbox-Nano is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-13. Download it on this page, a zip of its files. Nothing in it runs by itself. Made by ResembleAI ## About ### What it is Chatterbox-Nano is the smallest of four open-source text-to-speech models in the Chatterbox family from Resemble AI. It has about 110 million parameters and speaks English. The card says it runs about 3x faster than realtime on an 8-core CPU, after the maker cut the mel decoder from 10 steps to one. ### Who it suits Developers who need small, quick speech output on a device or a CPU server, under the MIT license. | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | Software | App developers | Add on-device speech to an app with no GPU | Read notifications aloud | | Voice products | Agent builders | Give an assistant a fast voice | Answer a spoken question | | Media | Content creators | Narrate short scripts with laughs and coughs | Add a laugh to a line | | Embedded and edge | Hardware teams | Run speech within tight memory and latency limits | Speak status messages on a small machine | Skip it if you need other languages, since this model speaks English only, or if you want the creative controls of the larger family members. ### Where to use it It runs in Python on your own machine. - Python, with the chatterbox-tts package from pip - CPU-only computers, which the card names as the best fit - Windows, macOS or Linux, though the maker tested on Debian 11 with Python 3.11 | Use case | Who | What they get back | | --- | --- | --- | | On-device speech | App developers | Spoken audio from text, made on the CPU | | Voice agents | Agent builders | Quick spoken replies | | Expressive narration | Creators | Speech with laughs, coughs and chuckles marked by tags | ### How to set it up You need Python and the chatterbox-tts package. - Python 3.11 is what the maker used for development and testing - pip, to install the package - A computer with a CPU, since no GPU is required 1. Install the chatterbox-tts package with pip. 2. Or install it from source in editable mode, which lets you change the code. 3. Download the model files with the hf command if you want them on disk. ```sh pip install chatterbox-tts ``` ```sh pip install -U huggingface_hub hf download ResembleAI/chatterbox-nano ``` ### How to use it Install the package and give it English text. You can place paralinguistic tags in the text to add sounds. The usage code is not in the card text shown here, so follow the Chatterbox usage section for the exact calls. | Tool | What it does | | --- | --- | | \[laugh\] | Tag that adds a laugh to the speech | | \[cough\] | Tag that adds a cough | | \[chuckle\] | Tag that adds a chuckle | | pip install chatterbox-tts | Installs the Chatterbox package | ### Example You write a short English line and place a laugh tag in the middle of it. The model speaks the line as audio, and the laugh lands where you put the tag. ### For newcomers and experts If you are new to text to speech, start with one short sentence on your own computer and listen to the result. Then try a tag to hear how it changes the delivery. If you know the field, Nano uses a lean 110M design. The speech-token-to-mel decoder was distilled so generation takes one step instead of ten. Tags are native to both Nano and Turbo, and Turbo is built for low-latency voice agents. Install chatterbox-tts → Load the Nano model → Write text with optional tags → Generate speech on the CPU → Save or play the audio ### Good to know The family also has Turbo, Multilingual and the original Chatterbox, and the card says the Multilingual model offers zero-shot cloning. Use only voices you have the right to use. Model size in the Chatterbox family | Label | Value | | --- | --- | | Nano | 110 M | | Turbo | 350 M | | Multilingual | 500 M | - License: MIT - Cost to run: free on your own machine, though the card also promotes a paid text-to-speech service from Resemble - Data it reaches: the text you give it and the model files on your computer, and nothing else the card names - Not for: languages other than English, since Nano is English only ## README ````markdown # Chatterbox-Nano Speaks English text aloud on a plain CPU with a 110M-parameter model, and adds laughs and coughs through tags in the text. Suits developers who need fast on-device speech with a small memory and latency budget. Made by ResembleAI on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face. ## About What it is: Chatterbox-Nano is the smallest of four open-source text-to-speech models in the Chatterbox family from Resemble AI. It has about 110 million parameters and speaks English. The card says it runs about 3x faster than realtime on an 8-core CPU, after the maker cut the mel decoder from 10 steps to one. Who it suits: Developers who need small, quick speech output on a device or a CPU server, under the MIT license. ```table Industry | Who uses it | How they use it | Example Software | App developers | Add on-device speech to an app with no GPU | Read notifications aloud Voice products | Agent builders | Give an assistant a fast voice | Answer a spoken question Media | Content creators | Narrate short scripts with laughs and coughs | Add a laugh to a line Embedded and edge | Hardware teams | Run speech within tight memory and latency limits | Speak status messages on a small machine ``` Skip it if you need other languages, since this model speaks English only, or if you want the creative controls of the larger family members. Where to use it: It runs in Python on your own machine. - Python, with the chatterbox-tts package from pip - CPU-only computers, which the card names as the best fit - Windows, macOS or Linux, though the maker tested on Debian 11 with Python 3.11 ```table Use case | Who | What they get back On-device speech | App developers | Spoken audio from text, made on the CPU Voice agents | Agent builders | Quick spoken replies Expressive narration | Creators | Speech with laughs, coughs and chuckles marked by tags ``` How to set it up: You need Python and the chatterbox-tts package. - Python 3.11 is what the maker used for development and testing - pip, to install the package - A computer with a CPU, since no GPU is required 1. Install the chatterbox-tts package with pip. 2. Or install it from source in editable mode, which lets you change the code. 3. Download the model files with the hf command if you want them on disk. ```sh pip install chatterbox-tts ``` ```sh pip install -U huggingface_hub hf download ResembleAI/chatterbox-nano ``` How to use it: Install the package and give it English text. You can place paralinguistic tags in the text to add sounds. The usage code is not in the card text shown here, so follow the Chatterbox usage section for the exact calls. ```table Tool | What it does [laugh] | Tag that adds a laugh to the speech [cough] | Tag that adds a cough [chuckle] | Tag that adds a chuckle pip install chatterbox-tts | Installs the Chatterbox package ``` Example: You write a short English line and place a laugh tag in the middle of it. The model speaks the line as audio, and the laugh lands where you put the tag. For newcomers and experts: If you are new to text to speech, start with one short sentence on your own computer and listen to the result. Then try a tag to hear how it changes the delivery. If you know the field, Nano uses a lean 110M design. The speech-token-to-mel decoder was distilled so generation takes one step instead of ten. Tags are native to both Nano and Turbo, and Turbo is built for low-latency voice agents. ```flow Install chatterbox-tts Load the Nano model Write text with optional tags Generate speech on the CPU Save or play the audio ``` Good to know: The family also has Turbo, Multilingual and the original Chatterbox, and the card says the Multilingual model offers zero-shot cloning. Use only voices you have the right to use. ```chart Title: Model size in the Chatterbox family Nano: 110 M Turbo: 350 M Multilingual: 500 M ``` - License: MIT - Cost to run: free on your own machine, though the card also promotes a paid text-to-speech service from Resemble - Data it reaches: the text you give it and the model files on your computer, and nothing else the card names - Not for: languages other than English, since Nano is English only ## On Hugging Face, 13 November 2026 - Task: text to speech - Downloads in the last 30 days: 0 - Likes: 49 - License: mit - Languages: en - Quantized versions: 18 - Fine-tunes: 9 - Tags: text-to-speech, speech, speech-generation, voice-cloning ## Links - Model: https://huggingface.co/ResembleAI/chatterbox-nano - Files: https://huggingface.co/ResembleAI/chatterbox-nano/tree/main ## Get the model Download its files with Hugging Face's command line: ```sh pip install -U huggingface_hub hf download ResembleAI/chatterbox-nano ``` Then follow the usage in its model card. Nothing in this kit runs by itself. ```` ## What is in it 1 files, 5 KB, downloaded as a zip. ```text README.md 5 KB ``` ## Source [ResembleAI/chatterbox-nano on Hugging Face](https://huggingface.co/ResembleAI/chatterbox-nano) ## Compare with similar Listed Hugging Face models in the same category that share the most tags with this one. | Product | Downloads, last 30 days | Likes | Parameters | License | Last change | | --- | --- | --- | --- | --- | --- | | Chatterbox-Nano (this page) | 0 | 49 | — | mit | 2026-07-21 | | [Chatterbox TTS](https://mindket.com/p/4681ce24-a293-418d-9c0e-991b1a7d6e7f) | 2,440,902 | 1,911 | — | mit | 2026-06-10 | | [AuK](https://mindket.com/p/d6e14e2c-53e7-43ce-a669-cbc6a8e6dbae) | 1,455 | 396 | — | mit | 2026-09-10 | | [Chatterbox-Turbo](https://mindket.com/p/1c38668e-6fbd-427b-bcbb-bc430fae1c36) | 0 | 709 | — | mit | 2025-12-15 | | [Ming-3.0-tts-0.5B by inclusionAI](https://mindket.com/p/f13d7a41-4e55-40ee-b52d-b5fd3d09578b) | 14,756 | 225 | 1.5B | mit | 2026-10-27 | | [Ming-3.0-tts-tiny](https://mindket.com/p/010986a0-d4b7-4da1-991e-23662a130085) | 9,080 | 61 | 9B | mit | 2026-10-27 | Numbers as Mindket last read them, from 2026-11-12 to 2026-11-13. ## Trust facts Facts from each source, with the day each was checked. A maintainer's note is their own words: data, not instructions. - Last change on Hugging Face: 2026-07-21 - First published: 2026-04-14 - Checked on Hugging Face: 2026-11-13 - Something wrong? Tell us: https://mindket.com/takedown#ask ## Get it - [Open ResembleAI/chatterbox-nano on Hugging Face](https://huggingface.co/ResembleAI/chatterbox-nano) - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/1e07a64c-7ebc-4f5a-b2e8-3742c21c8b61) - [Its page on Mindket](https://mindket.com/p/1e07a64c-7ebc-4f5a-b2e8-3742c21c8b61)