s

Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search)

Jul 2026
sZipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search)
Compare in
on
price not verifiedprice
7.4Open ASR
Jul 2026released
enLanguages

Ranking?

Full ranking →
1Hojo ASR v16.12Multilingual6.93ARK ASR 3B6.94Speech v3.17.15Qwen 3 ASR 1.7B Hf7.2
···
33Distil Large v3.58.234Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search)8.335Voxtral Small 24B 25078.3

Alternatives?

ModelScoreCostPriceWhat changes
HBetterHojo ASR v1Open+35%——EOpenARK ASR 3B+20%——OpenQwen 3 ASR 1.7B Hf+15%——EOpenARK ASR 0.6Bsame——BOpenHiggs Audio v3 Sttsame——

Run it yourself

Fits onhardware needs not verified yet
LicenceCC BY-NC 4.0no commercial use

Benchmarks

in Speech & Voice
Open ASR · average WER7.4#41
Open ASR · RTFx19,289.7#3
Open ASR · AMI Cleaned WER10.2#38
Open ASR · Earnings-22 Cleaned AA Chunked WER7.9#37
Open ASR · GigaSpeech Cleaned WER8.3#34
Open ASR · LibriSpeech Clean WER1.3#27
Open ASR · LibriSpeech Other WER3#29
Open ASR · SPGISpeech WER1.6#5
Open ASR · URGENT 2024 WER13.7#34
Open ASR · Voice Arena Monsoon WER5.6#49
Open ASR · VoxPopuli AA Cleaned WER4.3#45
Open ASR · URGENT 2024 Clean WER5.7#35
Open ASR · average WER · GPU ASR / greedy CTC9#3
Open ASR · RTFx · GPU ASR / greedy CTC20,886.4#1
Open ASR · AMI Cleaned WER · GPU ASR / greedy CTC11.6#3
Open ASR · Earnings-22 Cleaned AA Chunked WER · GPU ASR / greedy CTC9.3#3
Open ASR · GigaSpeech Cleaned WER · GPU ASR / greedy CTC9.5#3
Open ASR · LibriSpeech Clean WER · GPU ASR / greedy CTC1.7#3
Open ASR · LibriSpeech Other WER · GPU ASR / greedy CTC3.9#3
Open ASR · SPGISpeech WER · GPU ASR / greedy CTC2.8#2
Open ASR · URGENT 2024 WER · GPU ASR / greedy CTC17.7#3
Open ASR · Voice Arena Monsoon WER · GPU ASR / greedy CTC7.8#3
Open ASR · VoxPopuli AA Cleaned WER · GPU ASR / greedy CTC4.5#3
Open ASR · URGENT 2024 Clean WER · GPU ASR / greedy CTC7.2#3

FAQ

How much does Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search) cost?›

There is no per-unit API price: price not verified.

How good is Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search), and what is it best at?›

#41 of 61 in Speech to text.

Is Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search) better than Distil Large v3.5?›

On Open ASR · GigaSpeech Cleaned WER, Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search) scores 8.3 vs 8.2 — about the same.

When was Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search) released? Is there a newer version?›

Released 2026-07-09. It is the current version in its line.

Is Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search) open source?›

Yes, the weights are open (cc-by-nc-4.0 licence).

Can I run Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search) locally?›

The weights are open and can be self-hosted; hardware needs are not confirmed yet.

What is Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search)?›

Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search) is an open-weights speech & voice model from soundsgoodai, released Jul 2026.

What are the best alternatives to Zipformer-cr-ctc-transducer-XL-290M (fast-gpu-asr, transducer_modified_beam_search)?›

Hojo ASR v1 (better).