# Low CER

A Teochew (Chaoshan) orthographic recognition model fine-tuned based on Whisper-medium, supporting multi-dialect accent orthographic transcription

A Thai and English optical character recognition model fine-tuned from the TrOCR base handwriting model, excelling in processing handwritten text line images

This model is designed to recognize handwritten text from text line images in table cells, particularly suitable for handwritten text recognition in Finnish death records and census records from the 1930s.

An automatic speech recognition model based on Wav2Vec2-Base architecture, specifically fine-tuned for phoneme recognition on the LJSpeech Phonemes dataset

Fine-tuned based on Facebook's wav2vec2-base model, specifically optimized for Korean speech recognition, and can accurately transcribe Korean speech into text.

A specialized image-to-text model for the financial domain, supporting English text recognition, primarily used for processing image content in financial documents.

A model focused on Vietnamese speech recognition, supporting high-accuracy speech-to-text conversion.

A fine-tuned printed text recognition model based on microsoft/trocr-base-printed, mainly used for captcha recognition tasks

This model is a Japanese speech recognition model fine-tuned based on openai/whisper-small, supporting Japanese speech-to-text tasks.

Transformer-based French printed text OCR model, filling the gap of French version in TrOCR models

A phoneme recognition model based on the Wav2Vec2 architecture, fine-tuned on the LJSpeech Phonemes dataset, used to convert speech into phoneme sequences

A Cantonese speech recognition model fine-tuned based on OpenAI Whisper-small, achieving a CER of 7.93 on the Common Voice 16.0 test set

This is a large Conformer-Transducer model for transcribing Mandarin speech, with approximately 120 million parameters, trained on the AISHELL-2 dataset.

This is a non-autoregressive Citrinet model for Mandarin Chinese automatic speech recognition (ASR), with approximately 140 million parameters, using character encoding and CTC loss/decoding.

An Estonian automatic speech recognition model fine-tuned based on facebook/wav2vec2-xls-r-300m, trained with approximately 800 hours of diverse data

This is a Portuguese automatic speech recognition model trained on the Mozilla Common Voice 7.0 dataset, supporting Portuguese speech-to-text tasks.

This is an automatic speech recognition (ASR) model fine-tuned on the Common Voice 7.0 Thai dataset based on the wav2vec2-large-xlsr-53 model.

A fine-tuned Bengali automatic speech recognition model based on facebook/wav2vec2-xls-r-300m, trained using the OPENSLR_SLR53 dataset

A speech recognition model fine-tuned on the Common Voice 8.0 Czech dataset based on facebook/wav2vec2-xls-r-300m

A Chinese (zh-CN) speech recognition model fine-tuned on the Common Voice dataset based on facebook/wav2vec2-large-xlsr-53

This model is an automatic speech recognition model fine-tuned on the Romanian Common Voice 7.0 dataset based on facebook/wav2vec2-xls-r-1b.

An automatic speech recognition model fine-tuned on Bulgarian language datasets based on facebook/wav2vec2-xls-r-300m

This is a French automatic speech recognition model fine-tuned on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - FR dataset based on facebook/wav2vec2-xls-r-1b.

This is an automatic speech recognition (ASR) model fine-tuned on the Slovenian language (Common Voice 8.0) dataset based on facebook/wav2vec2-xls-r-300m.

An automatic speech recognition model fine-tuned on the Norwegian written language (Bokmål) speech dataset based on the facebook/wav2vec2-xls-r-1b model