NeuTTS-2E
Compact open-source text-to-speech model for controllable emotion (seven emotions, English), run on-device on phones, laptops and edge boards.
Find alternatives to NeuTTS-2E
What it does
- Text to speech
Features 3
- Seven emotion settingsOffers seven emotion settings: neutral, happy, sad, angry, surprised, fearful and disgusted
- Same voice, different emotionProduces different emotional deliveries while keeping the chosen speaker's voice
- On-device deploymentDeploys in GGML on phones, laptops, Raspberry Pi and Jetson Orin Nano
Sources
- Pricing
-
No pricing page or figure found on the product page or the site pages read 2026-11-06; the model is the contact route the page offers.
- Description
-
Company disclosure · 6 Nov 2026
Page read in a browser 2026-11-06; page: "Our compact open-source TTS model for controllable emotion", "under the NeuTTS Open License", ~125M active parameters, CPU, on-device; same kind of release as the recorded NeuTTS-2M. No price.
- What it does
-
Company disclosure · 6 Nov 2026
Page read in a browser 2026-11-06; page: "Our compact open-source TTS model for controllable emotion", "under the NeuTTS Open License", ~125M active parameters, CPU, on-device; same kind of release as the recorded NeuTTS-2M. No price.
- Sold within
-
Official documentation · 6 Nov 2026
Page read in a browser 2026-11-06; an open model obtained under its licence; it needs no parent product or account.
- Name
-
Company disclosure · 6 Nov 2026
Page read in a browser 2026-11-06; page: "Our compact open-source TTS model for controllable emotion", "under the NeuTTS Open License", ~125M active parameters, CPU, on-device; same kind of release as the recorded NeuTTS-2M. No price.
- URL
-
Company disclosure · 6 Nov 2026
Page read in a browser 2026-11-06; page: "Our compact open-source TTS model for controllable emotion", "under the NeuTTS Open License", ~125M active parameters, CPU, on-device; same kind of release as the recorded NeuTTS-2M. No price.
- Interfaces
-
Official documentation · 7 Nov 2026
Page: 'pip install neutts and start with sentences that give the model nothing'.
Also from Neuphonic
-
NeuTTS-2M Developer tool
NeuTTS-2M is a multilingual text-to-speech model with voice cloning that runs on local hardware.
-
Neuphonic API API service
Hosted real-time text-to-speech API with voice cloning, multilingual speech, timestamps and IPA pronunciation control, with Python and JavaScript SDKs.
Open Neuphonic in the directory
Something wrong here? Send a correction — quote this product id: neuphonic-neutts-2e.