Meddies ASR Synthetic Dialog Speech: Doctor-Patient Conversations

Synthetic doctor-patient dialog speech for automatic speech recognition training, generated with Fish Audio TTS. The dataset targets 200 hours of speech per language, using 75 English and 165 Chinese curated conversational voices. It was created by Meddies and last updated in July 2026.

Use Cases

• Train automatic speech recognition models based on synthetic medical dialog audio.
• Benchmark ASR performance on doctor-patient conversations based on the described multilingual speech data.
• Develop and test speech synthesis pipelines based on the curated voice registries mentioned.
• Fine-tune models for medical domain speech recognition based on the described conversational content.

Strengths

• Targets a substantial 200 hours of speech per language.
• Uses a curated registry of 75 English and 165 Chinese voices for diversity.
• Audio is generated at a professional 16 kHz mono FLAC quality.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Meddies

Collection Method

Synthetic generation using Fish Audio TTS (s2.1-pro-free) over curated conversational voice registries.

Freshness

Last updated 2026-07-03 05:38:40; freshness should be verified.