---
license: apache-2.0
language:
- lv
tags:
- Whisper
metrics:
- name: wer
  type: wer
  value: 9.56
pipeline_tag: automatic-speech-recognition
---

# Latvian Whisper small speech recognition model

Trained on combination of:
- Common Voice 17, custom selection of all validated clips, max 1000 clips per speaker
- Fleurs, test+train+validation

Regular whisper model and CTranslate2 converted version for use with [faster-whisper](https://github.com/SYSTRAN/faster-whisper) as part of [Home Assistant Whisper integration](https://www.home-assistant.io/integrations/whisper/) are available as well as GGML version for use with [whisper.cpp](https://github.com/ggerganov/whisper.cpp). 

For speech generation in Home Assistant use [Latvian Piper TTS voice](https://huggingface.co/RaivisDejus/Piper-lv_LV-Aivars-medium).

To improve speech recognition quality, more data is needed, donate your voice on [Balsu talka](https://balsutalka.lv/)