AUDIODIALOGUE
Natural English Dialogue Sets
Natural multi-speaker conversation with the turn structure, timing and speaker context needed for ASR, diarisation and conversational systems.
View datasetAUDIOSUPPORT DIALOGUE
Customer Service Dialogue Simulations
Support conversations are shaped by progression: the initial request, clarification, failed attempts, escalation and the eventual outcome. Customer Service Dialogue Simulations are designed around that sequence rather than treating each utterance as an independent example.
Scenario context and role information can remain attached to the dialogue, allowing a model to learn from how an issue develops across turns. Because the collection is scenario-based, it can target specific interaction patterns without implying access to private production call-centre recordings.
Technical structure
- Customer and agent role structure across a complete session
- Scenario context linked to turn-level transcript records
- Intent progression, escalation or outcome annotations where relevant
- Conversation-level metadata for grouping and analysis
Typical applications
Conversational AI, support agents, escalation modelling, intent progression, summarisation and outcome prediction.
Delivery note
Scenarios, roles and outcome structure are defined around the support behaviours the project needs to represent.
Delivery
The exact package depends on the collection scope. Where relevant, delivery can include task-specific records, manifests, provenance fields, stable identifiers, SHA-256 hashes, MLCommons Croissant 1.0 metadata and loading instructions.