The TTS pipeline loads two GGUFs: a MagpieTTS token generator and a NeMo NanoCodec decoder. The CLI downloads the complete default stack, including Magpie's tokenizer assets, with one command:
nemo-speech pull magpie nemo-speech synthesize "Hello from Magpie Multilingual." --output output.wavsynthesize performs the same verified pull automatically when its model options are omitted.
MagpieTTS token generator
Hugging Face:
nvidia/magpie_tts_multilingual_357m
# Download the v2602 GGUF and its matching tokenizer archive from their# immutable revisions. hf download nvidia/magpie_tts_multilingual_357m \ --include magpie_tts_multilingual_357m.v2602.f16.gguf \ --revision 452ef560f972c38d5fc16476259aac9456453547 \ --local-dir models/magpie-tts hf download nvidia/magpie_tts_multilingual_357m \ --include magpie_tts_multilingual_357m.nemo \ --revision 34d7e40da85cabc97f92198889b65cea27bc7fd1 \ --local-dir models/magpie-tts # Extract the tokenizer assets loaded by the runtime. mkdir -p models/magpie-tts/extracted tar -xf models/magpie-tts/magpie_tts_multilingual_357m.nemo \ -C models/magpie-tts/extractedMagpieTTS v2607 uses factor-2 frame stacking and must currently be converted locally before use:
hf download nvidia/magpie_tts_multilingual_357m \ magpie_tts_multilingual_357m.nemo \ --revision v2607 --local-dir models/magpie-tts-v2607 python3 convert_model.py models/magpie-tts-v2607/magpie_tts_multilingual_357m.nemo \ --outfile models/magpie-tts-v2607/magpie_tts_multilingual_357m.v2607.f16.gguf mkdir -p models/magpie-tts-v2607/extracted tar -xf models/magpie-tts-v2607/magpie_tts_multilingual_357m.nemo \ -C models/magpie-tts-v2607/extractedBoth v2602 (factor 1) and v2607 (factor 2) use the same NanoCodec decoder.
Tokenizer. MagpieTTS's tokenizer assets live inside the .nemo archive - they are not part of the GGUF. The built-in pull extracts only the required, pinned tokenizer members and verifies each one. For a custom Magpie checkpoint, extract its .nemo archive and pass that directory as --tokenizer-dir or --tts.tokenizer-model-dir. The model-specific IPA/text tokenizer assets are loaded from this directory. The GGUF and extracted directory must come from the same model revision. The runtime recognizes the exact v2602 and v2607 tokenizer layouts from model_config.yaml and rejects unknown layouts or a profile mismatch at startup. Newly converted GGUFs record the profile explicitly; older v2602 and v2607 GGUFs are identified from their text-vocabulary and frame-stacking dimensions. In particular, v2602 uses the Hindi character tokenizer, while v2607 uses the bundled Hindi IPA dictionary and also changes tokenizer order, Japanese ASCII casing, Italian/Vietnamese tokenizer names, and the supported language set. Japanese tokenization requires a build with NEMO_SPEECH_TTS_WITH_JA=ON (disabled by default), which builds Open JTalk, MeCab, and the NAIST dictionary. Mandarin requires NEMO_SPEECH_TTS_WITH_ZH=ON (disabled by default) and additionally uses cppjieba plus pypinyin-compatible tables bundled with the native runtime. Those tables are stored in Git LFS, so run git lfs install once and git lfs pull before configuring this feature. No Python environment is needed when serving zh or zh-CN. Run git lfs pull in the checkout before docker build as well, because the build context does not include .git.
Text normalization. TTS can optionally run Sparrowhawk TN before Magpie tokenization. Build with -DNEMO_SPEECH_WITH_NORM=ON, install the WFST dependencies with scripts/build_itn_deps.sh, and pass a TN grammar directory such as models/tn_configs through --tts.tn-model-dir. The expected multilingual layout matches ASR ITN: immediate language-named children such as en/, fr/, and vi/, each containing tokenize_and_classify.far, verbalize.far, and optionally post_process.far. A direct single-language grammar directory and the older split classify/ and verbalize/ layout remain supported. See
for server, YAML, and offline runner examples.
NanoCodec decoder
Hugging Face:
nvidia/nemo-nano-codec-22khz-1.89kbps-21.5fps
(no tokenizer is needed for the codec decoder).
Pull it independently with nemo-speech pull nano-codec. Pulling magpie does this automatically because the two models must run together.
Run the default stack:
nemo-speech synthesize "Hello from Magpie Multilingual." --output output.wavConverting custom TTS checkpoints
The unified
entry point accepts compatible local .nemo archives and extracted NeMo checkpoints. It defaults to --outtype f16 for MagpieTTS and NanoCodec; pass --outtype f32 to retain full precision. The converter is a source-tree Python tool and is not included in native release archives; see
for environment setup.
python3 convert_model.py custom-magpie.nemo --outfile custom-magpie.f16.ggufConversion does not require nemo_toolkit. The optional scripts/tts/tokenize-magpietts.py debugging helper does.
Once converted, point the server at them; see
.