---

license: cc-by-4.0 language:

• en
• bg
• hr
• cs
• da
• nl
• et
• fi
• fr
• de
• el
• hu
• it
• lv
• lt
• mt
• pl
• pt
• ro
• sk
• sl
• es
• sv
• ru
• uk library_name: transformers pipeline_tag: automatic-speech-recognition base_model:
• nvidia/parakeet-tdt-0.6b-v3
• moondream/parakeet-ultra tags:
• speech-recognition
• automatic-speech-recognition
• parakeet
• tdt
• fastconformer
• transformers
• nemo

---

🎙️ Fluid Speech Pro: Fast Multilingual Speech-to-Text

| | |

Description

Fluid Speech Pro is a 0.63-billion-parameter speech recognition model for fast, accurate speech-to-text, from the team behind FluidVoice. It is built on NVIDIA's Parakeet TDT 0.6B v3 and keeps its 25 European languages. It detects the language on its own, with no prompt needed.

It is the #2 open model on the Open ASR Leaderboard, with a 6.06 average word error rate. It is the most accurate open model under 1B parameters, #2 among open models on noisy speech, and it runs at about 3,700× real time on one GPU.

This model is ready for commercial and non-commercial use.

Key features

• Accurate: #2 open model on the Open ASR Leaderboard (6.06 average WER)
• Best under 1B: 10% fewer errors than the next open model under 1B parameters
• Strong in noise: #2 open model on URGENT 2024, the leaderboard's noisy-speech set (9.83 WER)
• Fast: about 3,700× real time on one GPU, so an hour of audio takes about a second
• Punctuation and capitalization included
• Word- and segment-level timestamps (NeMo)
• Clean mode (optional): edited text without repeated words and filler words (NeMo, ONNX and MLX)
• 25 European languages, detected automatically
• Five versions: NeMo and 🤗 Transformers, ONNX, MLX, GGUF and Core ML
• Permissive license: CC BY 4.0

Performance

Figure 1: Average word error rate (up is better) against model size for open models on the Open ASR Leaderboard. Fluid Speech Pro is the only top-5 open model under 1B parameters.

Figure 2: The best open models at 1B parameters and under, ranked by average word error rate (lower is better).

Figure 3: Word error rate on URGENT 2024, the leaderboard's noisy-speech set (lower is better).

Use case

For developers, researchers and companies building anything that turns speech into text: dictation, meeting notes, captions and subtitles, voice assistants, call and media analytics, and transcription services.

Release date

Hugging Face, November 2026.

Model architecture

Architecture type: FastConformer-TDT

Network architecture:

• A FastConformer encoder and a TDT (token-and-duration transducer) decoder, as in Parakeet TDT 0.6B v3
• One small extra block between the encoder's subsampling and its first Conformer layer
• 0.63 billion parameters

Input

• Input type: audio, 16 kHz, mono
• Input format: .wav, .flac and other formats your audio loader reads

Output

• Output type: text, with punctuation and capitalization
• Optional: word and segment timestamps (NeMo)

All versions

Version	Best for	Clean mode
NeMo + 🤗 Transformers (this one)	NVIDIA GPUs and servers; the reference model	Yes (NeMo)
ONNX	CPUs, with onnx-asr or sherpa-onnx	Yes
MLX	Apple Silicon Macs, with parakeet-mlx	Yes
GGUF	C++ with parakeet.cpp, CPU or GPU, no Python	No
Core ML	Apple apps in Swift, on the Neural Engine, with FluidAudio	Not yet

How to use this model

🤗 Transformers

Needs Transformers 5.22 or newer:

pip install -U transformers

from transformers import pipeline

asr = pipeline("automatic-speech-recognition", model="FluidVoice-Labs/fluid-speech-pro")
print(asr("audio.wav")["text"])

The Transformers weights leave out the model's one small extra block and decode greedily. In our tests they give nearly the same results as the NeMo model.

NVIDIA NeMo

The NeMo checkpoint is the reference model and adds timestamps and clean mode. It loads through fluid_speech.py from this repository:

pip install "nemo_toolkit[asr]"
hf download FluidVoice-Labs/fluid-speech-pro fluid_speech.py --local-dir .

import fluid_speech

model = fluid_speech.load("FluidVoice-Labs/fluid-speech-pro", map_location="cuda")

# Transcribe
print(model.transcribe(["audio.wav"])[0].text)

# With word and segment timestamps
out = model.transcribe(["audio.wav"], timestamps=True)[0]
for word in out.timestamp["word"]:
    print(f"{word['start']:.2f}s - {word['end']:.2f}s: {word['word']}")

Clean mode

By default the model writes exactly what was said. Clean mode gives edited text instead: it drops repeated words ("I I think") and filler words ("um", "uh"). Use it for notes, captions and dictation, and leave it off when you need a word-for-word record.

clean = fluid_speech.load("FluidVoice-Labs/fluid-speech-pro", map_location="cuda", clean_mode=True)
print(clean.transcribe(["audio.wav"])[0].text)

Decoding: greedy or beam search

The model decodes greedily by default (beam 1). The leaderboard numbers use beam search with 4 beams, which makes about 0.6% fewer errors on average. On a GPU with batching, 1, 2 and 4 beams run at about the same speed; greedy does the least work per clip and is a little better on very noisy audio.

Beams	Errors vs greedy	Speed on a batched GPU
1 (default)	baseline	baseline
2	0.3% fewer	about the same
4	0.6% fewer	about the same

To decode like the leaderboard:

model.cfg.decoding.strategy = "malsd_batch"
model.cfg.decoding.beam.beam_size = 4
model.change_decoding_strategy(model.cfg.decoding)

Software integration

• Runtimes: 🤗 Transformers, NVIDIA NeMo, ONNX Runtime (onnx-asr, sherpa-onnx), MLX (parakeet-mlx), parakeet.cpp (GGUF), Core ML (FluidAudio)
• Hardware: NVIDIA GPUs for full speed; it also runs on CPU and Apple Silicon
• Operating systems: Linux, macOS, Windows

Model version

Fluid Speech Pro 1.0

Evaluation datasets

The Open ASR Leaderboard's public test sets: AMI (cleaned), Earnings-22 (cleaned, chunked), GigaSpeech (cleaned), SPGISpeech, URGENT 2024 (noisy speech), Voice Arena Monsoon (Indian English) and VoxPopuli (cleaned).

Results

Open ASR Leaderboard

Open models on the Open ASR Leaderboard: average word error rate over its seven public and two private test sets (lower is better):

Model	Parameters	Average WER
Musci-ASR	2B	5.70
Fluid Speech Pro	0.63B	6.06
NVIDIA Nemotron 3.5 Transcribe	30B	6.13
Qwen3-ASR	1.7B	6.15
IBM Granite Speech 5.0	2B	6.68
Cohere Transcribe	2B	6.72
NVIDIA Parakeet TDT 0.6B v2	0.6B	6.93
NVIDIA Parakeet TDT 0.6B v3	0.6B	7.04
Fermion Research Phonon-2	0.6B	7.86

Results by test set

Word error rate (%) on the leaderboard's seven public test sets:

Model	Public average	AMI	Earnings-22	GigaSpeech	SPGISpeech	URGENT 2024	Monsoon	VoxPopuli
Musci-ASR	5.15	6.53	5.83	6.41	1.86	10.49	3.44	1.47
Fluid Speech Pro	5.43	7.55	5.42	7.51	2.31*	9.83	3.71	1.70
NVIDIA Nemotron 3.5 Transcribe	5.51	7.78	5.25	7.18	1.91	10.94	3.63	1.89
Qwen3-ASR	5.75	8.31	5.84	7.22	2.57	9.93	3.50	2.85
IBM Granite Speech 5.0	6.05	6.96	6.22	8.14	3.36	10.22	3.94	3.52
Cohere Transcribe	6.26	7.02	7.88	7.88	2.74	9.50	3.44	5.38
NVIDIA Parakeet TDT 0.6B v2	6.43	9.09	6.76	8.15	1.94	11.42	3.89	3.78
NVIDIA Parakeet TDT 0.6B v3	6.51	9.42	5.85	7.99	3.63	11.52	4.14	3.05
Fermion Research Phonon-2	7.23	9.32	6.96	8.35	3.70	14.90	4.91	2.46

* In clean mode, which matches SPGISpeech's edited references; the leaderboard allows one transcription mode per test set.

Other languages

Fluid Speech Pro supports 25 languages. On the Open ASR Leaderboard's multilingual track it makes 12% fewer errors than Parakeet TDT 0.6B v3, and it is better in all seven of the track's languages:

Language	Parakeet TDT 0.6B v3	Fluid Speech Pro	Fewer errors
German	4.12	3.76	9%
French	5.40	5.10	6%
Italian	4.73	4.16	12%
Spanish	3.71	3.38	9%
Portuguese	6.08	4.88	20%
Dutch	7.53	6.82	9%
English	7.04	6.06	14%
Average	5.52	4.88	12%

Word error rate (%) on the Open ASR Leaderboard's multilingual track, lower is better.

It also got better than Parakeet TDT 0.6B v3 in every one of the 24 languages beyond English. On FLEURS, scored the way the leaderboard scores its multilingual track, it makes nearly 20% fewer errors on average.

Figure 4: Word error rate on FLEURS for Parakeet TDT 0.6B v3 (grey) and Fluid Speech Pro (blue), lower is better.

Language	Parakeet TDT 0.6B v3	Fluid Speech Pro	Fewer errors
Estonian	13.28	9.51	28%
Latvian	21.76	16.39	25%
Italian	2.48	1.91	23%
Lithuanian	21.04	16.21	23%
Maltese	19.23	15.05	22%
Hungarian	13.58	10.65	22%
Slovak	9.18	7.20	22%
Finnish	11.19	8.82	21%
Czech	11.06	8.83	20%
Portuguese	4.47	3.57	20%
Slovenian	22.48	17.97	20%
Ukrainian	5.95	4.77	20%
Greek	21.16	17.10	19%
Romanian	11.38	9.28	18%
Spanish	3.22	2.66	17%
Swedish	13.39	11.10	17%
Polish	6.84	5.70	17%
German	4.18	3.52	16%
Danish	17.20	14.57	15%
Bulgarian	11.94	10.19	15%
Croatian	11.40	9.73	15%
Dutch	6.39	5.51	14%
Russian	4.92	4.35	12%
French	4.75	4.65	2%
Average of all 24	11.35	9.13	20%

Word error rate (%) on FLEURS test sets, greedy decoding. A future version will be trained on more multilingual data.

Limitations

• Please check the model on your own audio, accents and languages before you rely on it.
• Like every speech model, it makes more mistakes on heavy noise, overlapping speakers, strong accents it has rarely heard, and rare names and terms.
• Clean mode edits the text. Don't use it where you need an exact record.

License

CC BY 4.0. If you use the model, please credit it as Fluid Speech Pro by FluidVoice.

Built on NVIDIA Parakeet TDT 0.6B v3 and Moondream Parakeet Ultra, both CC BY 4.0.

Citation

@misc{fluidspeechpro2026,
  title  = {Fluid Speech Pro},
  author = {{FluidVoice}},
  year   = {2026},
  url    = {https://huggingface.co/FluidVoice-Labs/fluid-speech-pro}
}

References

1. Parakeet TDT 0.6B v3, NVIDIA
2. Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition
3. Efficient Sequence Transduction by Jointly Predicting Tokens and Durations
4. Open ASR Leaderboard, Hugging Face