# Correct Wav2Vec2 ASR output

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcorrect-wav2vec2-asr-output%2F66589%3E%5C
**Category:** Beginners\
**Created:** [December 21, 2023, 10:07am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcorrect-wav2vec2-asr-output%2F66589 "2023-12-21T10:07:52Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Davide85](https://avatars.discourse-cdn.com/v4/letter/d/67e7ee/32.png) [@Davide85](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FDavide85)%5C
**Post date:** [December 21, 2023, 10:07am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcorrect-wav2vec2-asr-output%2F66589%2F1 "2023-12-21T10:07:53Z")

</div>

I have fine-tuned Wav2Vec2-960h on my custom Italian speech dataset. At the inference, I have some transcription errors. Can I fix them using a LLM? Any idea or practical example?

Thanks in advance,

Davide