Common Voice 25: Scripted Speech Audio in Chinese (Taiwan)

Mozilla Common Voice Scripted Speech 25.0 provides audio data for the Chinese (Taiwan) language. This mirror by OpenFormosa embeds the audio into Parquet files at a 48 kHz sampling rate. The dataset was last updated on Hugging Face on 2026-06-06.

Use Cases

• Training automatic speech recognition (ASR) models based on scripted speech audio.
• Fine-tuning text-to-speech (TTS) synthesis models using high-quality 48 kHz audio samples.
• Benchmarking speech model performance on the Chinese (Taiwan) language subset of Common Voice.
• Studying phonetic and prosodic features of Taiwanese Mandarin from recorded scripted prompts.

Strengths

• Audio is provided at a high 48 kHz sampling rate, which may preserve detailed acoustic information.
• Data originates from the Mozilla Common Voice project, a known open-source speech data initiative.

Limitations

• Description metadata is limited; actual data quality, size, and column structure require manual inspection after download.
• Row count, license, and detailed file formats are unknown, which may limit suitability assessment.
• Last updated 2026-06-06 17:44:08; freshness should be verified against the original Mozilla source.

Provenance

Source

Mozilla Data Collective - Common Voice Scripted Speech 25.0 - Chinese (Taiwan)

Collection Method

Mirrored and embedded into Parquet format by OpenFormosa.

Freshness

Last updated on the platform: 2026-06-06 17:44:08.

Geography

Taiwan (Chinese (Taiwan) language)