| --- |
| language: |
| - en |
| task_categories: |
| - automatic-speech-recognition |
| - text-to-speech |
| - audio-classification |
| pretty_name: British English Full-Duplex Two-Speaker Conversational Dataset (Prompt-Based Sample) |
| tags: |
| - audio |
| - speech |
| - conversational |
| - multi-speaker |
| - full-duplex |
| - turn-taking |
| - prompt-based |
| - open-topic |
| - british-english |
| - english |
| - sample |
| size_categories: |
| - n<1K |
| extra_gated_prompt: >- |
| This is a free prompt-based preview sample of OcularAI's full British English |
| Full-Duplex Two-Speaker Conversational Dataset. It contains high-fidelity |
| recordings of identifiable human speakers and self-reported demographic |
| attributes. By requesting access you agree to use the data for evaluation, |
| research, and AI development only, to follow the terms of use, and to make no |
| attempt to identify or contact any speaker. For commercial licensing or access |
| to the full dataset, contact OcularAI. |
| extra_gated_fields: |
| Full name: text |
| Company / Affiliation: text |
| Work email: text |
| Intended use: text |
| Interested in the full dataset: checkbox |
| I agree to the data use terms: checkbox |
| --- |
| |
| # British English Full-Duplex Two-Speaker Conversational Dataset — Prompt-Based Sample |
|
|
| > **This is a free prompt-based preview** of OcularAI's full British English |
| > Full-Duplex Two-Speaker Conversational Dataset. Each conversation is an |
| > **open, natural discussion** — paired speakers were prompted to talk freely on |
| > a topic of their choice. Audio is **high-fidelity FLAC**, with each speaker on |
| > an independent, isolated track. (For conversations *directed* by a designed |
| > scenario targeting a specific turn-taking / voice-dynamics behavior, see the |
| > companion **Scenario-Based** samples in the same collection.) |
| > **For full-dataset access or commercial licensing, contact OcularAI.** |
|
|
| ## What's in this sample |
|
|
| - **28 conversations** of natural British-English full-duplex dialogue |
| - **56 isolated speaker tracks** — one per speaker, lossless **FLAC** |
| - All speakers based in the **United Kingdom** (self-reported), British-accented English |
| - Self-reported speaker demographics (gender, location, ethnicity) |
|
|
| ## Dataset summary |
|
|
| Natural, **unscripted, two-speaker British-English conversations** recorded by |
| fluent UK-based English speakers. Each session is a spontaneous discussion |
| between a matched pair of speakers on an everyday topic of their choice. |
|
|
| The recordings support the development of next-generation AI systems — helping |
| them better understand **natural speech patterns, conversational flow, |
| turn-taking, and real-world human interaction**. Each speaker is captured on an |
| **independent, isolated audio track**, enabling per-speaker analysis, |
| diarization, full-duplex modeling, ASR, and TTS. |
|
|
| ## Dataset structure |
|
|
| **One row per speaker track**, ordered by `room_name` so a conversation's two |
| speakers sit adjacent. Each row is one isolated voice with its own metadata; join |
| on `room_name` to reconstruct the conversation. Audio is lossless FLAC. |
|
|
| ### Fields (per row = one speaker's track) |
|
|
| | field | description | |
| |-------|-------------| |
| | `file_name` | this speaker's isolated FLAC track | |
| | `room_name` | conversation/session key — shared by both speakers | |
| | `conversation_id` | conversation identifier | |
| | `slot_number` | recording slot index | |
| | `role` | `SPEAKER_A` or `SPEAKER_B` | |
| | `speaker_id` | stable speaker identifier | |
| | `duration_seconds` | track duration | |
| | `sample_rate_hz` | audio sample rate | |
| | `bit_depth` | audio bit depth | |
| | `language` | spoken language (`en-GB`) | |
| | `accent` | `British` | |
| | `prompt` | the open-topic instruction given to the pair | |
| | `gender` | self-reported | |
| | `city`, `country` | self-reported location | |
| | `ethnicity` | self-reported | |
| | `fluent_languages` | languages the speaker is fluent in | |
|
|
| Rows are ordered by `room_name` so a conversation's two speakers appear |
| adjacent. Join on `room_name` to reconstruct a full conversation. |
|
|
| ## The full dataset |
|
|
| This sample is a small, representative slice of OcularAI's full British English |
| Full-Duplex Two-Speaker Conversational Dataset. The complete corpus is |
| substantially larger and continues to grow, and spans both open-topic and |
| scenario-directed conversations. Contact OcularAI for full access and commercial |
| licensing terms. |
|
|
| ## Privacy & consent |
|
|
| Speaker **names and emails are removed**. Demographic fields are self-reported. |
| Recordings were collected from consenting, compensated participants for AI |
| research. Access requires agreeing to evaluation/research-only use and no |
| re-identification. |
|
|
| ## Audio note |
|
|
| Tracks are lossless **FLAC**. Decode with `datasets>=4.0` (torchcodec/FFmpeg), |
| `soundfile`, `librosa`, or any standard audio stack. |
|
|
| ## Licensing |
|
|
| © OcularAI. All rights reserved. This sample is provided for evaluation only. |
| Licensing terms — including full commercial usage rights — are by agreement. |
| Contact OcularAI for licensing. |
|
|
