---
title: "Build Low-Latency Multilingual Voice Agents: Open Weights & Full Deployment Control with NVIDIA Magpie TTS"
description: "NVIDIA Magpie Multilingual TTS is a 364M-parameter open-weights model for multilingual voice applications, supporting 12 languages, including newly added Modern Standard Arabic, Korean, and Brazilian Portuguese. It is positioned for cascaded voice-agent architectures in which ASR, TTS, and LLM components are independently deployed, tuned, and scaled on controlled infrastructure. Frame stacking predicts two audio frames per decoding step, while a local transformer models dependencies between simultaneously generated codebook tokens; NVIDIA reports 32–79 ms single-stream TTFA across its listed GPUs. Compared with the previous release, French and Spanish CER improved to 1.54% and 0.60%, while SSIM reached 0.747 and 0.793; German SSIM also increased to 0.742. The Hugging Face checkpoint supports research and fine-tuning, while NVIDIA NIM provides optimized production serving and NeMo supports customization."
---

# Build Low-Latency Multilingual Voice Agents: Open Weights & Full Deployment Control with NVIDIA Magpie TTS

[Hugging Face](https://yomu.fyi/company/hugging-face) · Maryam Motamedi, Mikyas Desta, Jason Li, Jason Roche · Aug 10, 2026

**Type:** Announcement

## Summary

NVIDIA Magpie Multilingual TTS is a 364M-parameter open-weights model for multilingual voice applications, supporting 12 languages, including newly added Modern Standard Arabic, Korean, and Brazilian Portuguese. It is positioned for cascaded voice-agent architectures in which ASR, TTS, and LLM components are independently deployed, tuned, and scaled on controlled infrastructure. Frame stacking predicts two audio frames per decoding step, while a local transformer models dependencies between simultaneously generated codebook tokens; NVIDIA reports 32–79 ms single-stream TTFA across its listed GPUs. Compared with the previous release, French and Spanish CER improved to 1.54% and 0.60%, while SSIM reached 0.747 and 0.793; German SSIM also increased to 0.742. The Hugging Face checkpoint supports research and fine-tuning, while NVIDIA NIM provides optimized production serving and NeMo supports customization.

## Context

Voice applications must support multiple languages while keeping latency low, meeting privacy and data-residency requirements, allowing pronunciation and voice customization, and providing predictable performance on infrastructure controlled by the developer.

## Approach / What changed

Magpie combines open weights, a 12-language multilingual speaker representation, IPA grapheme-to-phoneme processing, custom pronunciation dictionaries, frame stacking, and a local transformer. It can run as a Hugging Face checkpoint for research and fine-tuning or as an NVIDIA NIM for optimized serving, with NeMo available for domain customization.

## Takeaways

- Magpie supports Hindi and Japanese code-switching through IPA grapheme-to-phoneme processing and custom pronunciation dictionaries, helping handle names, technical terms, and mixed-language content.
- In the cited on-premises NIM measurements, B200 achieved 32 ms single-stream TTFA and 239 ms at 64 streams, with reported throughput of 319.81× and 319.81× real time respectively.
- The Nemotron Voice Agent developer example combines Nemotron Speech, Magpie TTS, Nemotron models, NVIDIA NIM, and NeMo for interruptible, multilingual, multimodal voice-agent deployments.

**Tags:** [Architecture](https://yomu.fyi/topic/architecture), [Deployment](https://yomu.fyi/topic/deployment), [Machine Learning](https://yomu.fyi/topic/machine-learning), [Open Source](https://yomu.fyi/topic/open-source), [Performance](https://yomu.fyi/topic/performance)

- Source: [Hugging Face](https://huggingface.co/blog/nvidia/magpie-tts-multilingual-voice-agents)
- Source URL: https://huggingface.co/blog/nvidia/magpie-tts-multilingual-voice-agents
- Ingested by Yomu: 2026-08-29T05:41:01.054Z

[Read original post](https://huggingface.co/blog/nvidia/magpie-tts-multilingual-voice-agents)