Speech Recognition On Common Voice Vi
Metriken
Test WER
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Modellname | Test WER | Paper Title | Repository |
|---|---|---|---|
| wav2vec2-base-vietnamese-160h (No Language Model) | 10.78 | Wav2vec2 Base Vietnamese 160h | |
| Vietnamese end-to-end speech recognition using wav2vec 2.0 by VietAI | 11.52 | Vietnamese end-to-end speech recognition using wav2vec 2.0 | - |
| khanhld/chunkformer-large-vie | 6.66 | ChunkFormer: Masked Chunking Conformer For Long-Form Speech Transcription | - |
0 of 3 row(s) selected.