Spc R Segmented: Diarized and Merged German Speech Segments

A processed speech dataset derived from i4ds/spc_r. Each row represents a merged speech segment from a single speaker, created by applying speaker diarization and merging consecutive segments from the same speaker. The dataset was created by i4ds and last updated on Hugging Face in February 2026.

Use Cases

• Training speaker diarization models based on pre-labeled speech segments.
• Developing automatic speech recognition systems using segmented and speaker-attributed audio.
• Analyzing speaker turn patterns and speech duration in conversational audio.
• Benchmarking audio processing pipelines that handle merging of consecutive speaker segments.

Strengths

• Processed with a defined pipeline using pyannote/speaker-diarization-3.1 for diarization.
• Segments are merged to reduce fragmentation, which may improve downstream processing.
• Last updated on 2026 02 25, suggesting recent maintenance.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect geographic or source bias inherent to the original i4ds/spc_r dataset.

Provenance

Source

i4ds

Collection Method

Derived from i4ds/spc_r audio and SRT subtitles via a diarization and merging pipeline.

Time Range

null

Freshness

Last updated 2026-02-25 15:35:45.

Geography

null