Recent studyBlind TTS Elo is live. Compare two anonymous voice samples, vote after listening, and help separate real preference signal from noise.Vote in the study ->
Codesota · Models · Canary-qwen-2.5b6 results · 6 benchmarks
Model card

Canary-qwen-2.5b.

unknown
§ 02 · Benchmarks

Every benchmark Canary-qwen-2.5b has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01LibriSpeechSpeech · Speech Recognitionwer3.1%#28/92source ↗
02GigaSpeechSpeech · Speech Recognitionwer9.4%#45/47source ↗
03Earnings-22Speech · Speech Recognitionwer10.4%#46/50source ↗
04TED-LIUMSpeech · Speech Recognitionwer2.7%#47/50source ↗
05VoxPopuliSpeech · Speech Recognitionwer5.7%#54/55source ↗
06SPGISpeechSpeech · Speech Recognitionwer1.9%#55/56source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area

Where Canary-qwen-2.5b actually performs.

Speech
6
benchmarks
avg rank #45.8
§ 04 · Papers

1 paper with results for Canary-qwen-2.5b.

  1. 2025-03-07· 6 results

    Training and Inference Efficiency of Encoder-Decoder Speech Models

§ 06 · Sources & freshness

Where these numbers come from.

pwc-dump
6
results
0 of 6 rows marked verified.