rhasspy/voice-incubator-models

Voice Incubator models

The models the Voice Incubator app downloads the first time it starts, as it loads them (Parakeet or its language's speech model, GLiNER), and those its image's build fetches (kaldi/, micro-wake-word/).

Folder	What	From	License
parakeet-tdt_ctc-110m-en-int8/	NVIDIA's Parakeet TDT-CTC 110M, English: the transducer (encoder, decoder, joiner, tokens.txt) with its encoder's MatMul weights quantized to int8 (onnxruntime's dynamic quantization), and the CTC branch's head (ctc_head.npz), so that one encoder gives both the CTC log-probabilities commands are matched on and the transcripts	nvidia/parakeet-tdt_ctc-110m, as exported by csukuangfj/sherpa-onnx-nemo-parakeet_tdt_transducer_110m-en-36000 (e9bea5a) and csukuangfj/sherpa-onnx-nemo-parakeet_tdt_ctc_110m-en-36000 (3af92f1)	CC-BY-4.0, © NVIDIA Corporation
stt_de_fastconformer_hybrid_large_pc-int8/	NVIDIA's FastConformer Hybrid Transducer-CTC Large, German: laid out as Parakeet's above (the transducer with its encoder's MatMul weights in int8, and the CTC head beside it), for the app in German	nvidia/stt_de_fastconformer_hybrid_large_pc, exported with NeMo 2.3.3 (MODEL_CARD.md)	CC-BY-4.0, © NVIDIA Corporation
stt_es_fastconformer_hybrid_large_pc-int8/	NVIDIA's FastConformer Hybrid Transducer-CTC Large, Spanish: laid out as Parakeet's above (the transducer with its encoder's MatMul weights in int8, and the CTC head beside it), for the app in Spanish	nvidia/stt_es_fastconformer_hybrid_large_pc, exported with NeMo 2.3.3 (MODEL_CARD.md)	CC-BY-4.0, © NVIDIA Corporation
stt_fr_fastconformer_hybrid_large_pc-int8/	NVIDIA's FastConformer Hybrid Transducer-CTC Large, French: laid out as Parakeet's above (the transducer with its encoder's MatMul weights in int8, and the CTC head beside it), for the app in French	nvidia/stt_fr_fastconformer_hybrid_large_pc, exported with NeMo 2.3.3 (MODEL_CARD.md)	CC-BY-4.0, © NVIDIA Corporation
stt_it_fastconformer_hybrid_large_pc-int8/	NVIDIA's FastConformer Hybrid Transducer-CTC Large, Italian: laid out as Parakeet's above (the transducer with its encoder's MatMul weights in int8, and the CTC head beside it), for the app in Italian	nvidia/stt_it_fastconformer_hybrid_large_pc, exported with NeMo 2.3.3 (MODEL_CARD.md)	CC-BY-4.0, © NVIDIA Corporation
stt_nl_fastconformer_hybrid_large_pc-int8/	NVIDIA's FastConformer Hybrid Transducer-CTC Large, Dutch: laid out as Parakeet's above (the transducer with its encoder's MatMul weights in int8, and the CTC head beside it), for the app in Dutch	nvidia/stt_nl_fastconformer_hybrid_large_pc, exported with NeMo 2.3.3 (MODEL_CARD.md)	CC-BY-4.0, © NVIDIA Corporation
stt_pl_fastconformer_hybrid_large_pc-int8/	NVIDIA's FastConformer Hybrid Transducer-CTC Large, Polish: laid out as Parakeet's above (the transducer with its encoder's MatMul weights in int8, and the CTC head beside it), for the app in Polish	nvidia/stt_pl_fastconformer_hybrid_large_pc, exported with NeMo 2.3.3 (MODEL_CARD.md)	CC-BY-4.0, © NVIDIA Corporation
stt_ru_fastconformer_hybrid_large_pc-int8/	NVIDIA's FastConformer Hybrid Transducer-CTC Large, Russian: laid out as Parakeet's above (the transducer with its encoder's MatMul weights in int8, and the CTC head beside it), for the app in Russian	nvidia/stt_ru_fastconformer_hybrid_large_pc, exported with NeMo 2.3.3 (MODEL_CARD.md)	CC-BY-4.0, © NVIDIA Corporation
stt_ua_fastconformer_hybrid_large_pc-int8/	NVIDIA's FastConformer Hybrid Transducer-CTC Large, Ukrainian: laid out as Parakeet's above (the transducer with its encoder's MatMul weights in int8, and the CTC head beside it), for the app in Ukrainian	nvidia/stt_ua_fastconformer_hybrid_large_pc, exported with NeMo 2.3.3 (MODEL_CARD.md)	CC-BY-4.0, © NVIDIA Corporation
gliner2.5-small-v1-onnx/	fastino's GLiNER 2.5 small for the app's command fallback: ONNX networks with the encoder's layers in 8 bits and its embedding in 4 (MatMulNBits, block 32), the classifier's weights and the tokenizer (gliner2_onnx.json says how)	fastino/gliner2.5-small-v1	Apache-2.0
kaldi/	The wake word verifier's Kaldi. en_US-rhasspy.tar.gz: the U.S. English nnet3 acoustic model and IPA lexicon of Rhasspy's Kaldi profile, without its G2P model and language model (the verifier never guesses a pronunciation). tools-x86_64.tar.gz and tools-aarch64.tar.gz: Kaldi's and OpenFst's binaries, libraries, headers and utils//steps/ scripts, which the image's build compiles the verifier's resident decoder against	rhasspy/en-us_kaldi-rhasspy; Kaldi and OpenFst 1.6.7 -- x86-64 built from source, aarch64 the Kaldi of 2021 that Rhasspy shipped (kaldi-manylinux)	MIT (the model); Apache-2.0 (Kaldi, OpenFst)
micro-wake-word/	The microWakeWord models the app pushes to satellites, with their settings: okay_nabu (multilingual) and stop as Kevin Ahrendt released them, vad, hey_jarvis and hey_mycroft as ESPHome publishes them	kahrendt/microWakeWord releases okay_nabu_20241226.3 and stop; esphome/micro-wake-word-models (models/v2/)	Apache-2.0

The app also downloads EmbeddingGemma 2 (onnx/model_q4.onnx and its tokenizer) from onnx-community/embeddinggemma-2-ONNX, and its image's build ReDimNet-b2 (speaker identification) from OpenVoiceOS/redimnet-b2-vox2-onnx.