Benchmarking Malaysian Speech-to-Text models, HuggingFace space at https://huggingface.co/spaces/mesolitica/malaysian-stt-leaderboard
📈 We evaluate models based on 3 datasets,
- Malaya-Speech test set, Malay language, https://huggingface.co/datasets/huseinzol05/malaya-speech-stt-test-set/tree/main/malaya-speech
- Fleurs MS-MY test set, Malay language, https://huggingface.co/datasets/huseinzol05/malaya-speech-stt-test-set/tree/main/fleurs-ms-my
- IMDA TTS first 700 audio files, English language but with Manglish slang, https://huggingface.co/datasets/mesolitica/IMDA-TTS
- We filtered test set that contain numbers because malaya-speech transducer trained on normalized numbers.
- We lower case because malaya-speech transducer trained on lower case.
- We removed punctuation because malaya-speech transducer trained without punctuation.