Europe · Bulgaria
Bulgarian conversational speech dataset
1,500 hours of spontaneous Bulgarian conversations recorded in Bulgaria across 150 vetted native speakers. Dual-channel audio with time-aligned transcripts, speaker metadata (age, gender, region, device), and full chain-of-consent documentation. Licensed for commercial ASR, TTS, and voice AI training.
Dataset specifications
| Language | Bulgarian |
|---|---|
| Recorded in | Bulgaria |
| Speech type | Spontaneous two-person conversations |
| Volume | 1,500 hours |
| Speakers | 150 vetted native speakers |
| Audio | Dual-channel (one channel per speaker) |
| Transcripts | Time-aligned, human-verified |
| Metadata | Age, gender, region, recording device |
| License | Commercial (training & evaluation) |
| Price | $85 per hour of audio |
Need more hours, another dialect or your own recording setup? We collect speech data to spec.
Built for
Training and evaluating speech models that must handle real Bulgarian speech:
- Automatic speech recognition (ASR / STT)
- Text-to-speech & voice synthesis prosody
- Voice agents and IVR for Bulgaria
- Speaker diarization & turn-taking models
- Speech LLM / audio-understanding fine-tuning
Why conversational data?
Models trained on read speech consistently underperform on spontaneous audio. Disfluencies, overlap, and natural speech rates aren’t in the training distribution. This dataset closes that gap for Bulgarian. Read more in our guide to conversational vs. scripted speech data.
Free samples
Audit the Bulgarian data yourself
Share your company email and we’ll email you representative audio, matching transcripts, and the metadata schema. Check audio quality, transcript accuracy, and speaker diversity before you spend a dollar.
Public alternatives
Public Bulgarian speech data vs. this dataset
| Source | What it is | License | Commercial training |
|---|---|---|---|
| MATERIAL Bulgarian-English Language Pack (LDC2024S07) (opens in a new tab)Linguistic Data Consortium (LDC) | About 78 hours of Bulgarian conversational telephone speech with English translations, annotations and queries; transcripts cover about 40% of the files. | LDC MATERIAL Bulgarian-English Agreement (For-Profit, Non-Member or Not-For-Profit); fee applies | Paid license |
| VoxPopuli (opens in a new tab)Meta (facebookresearch) | European Parliament event recordings from 2009 to 2020; Bulgarian is unlabeled audio only, without transcripts. | CC0 for the VoxPopuli data; the raw audio falls under the European Parliament legal notice | Unclear |
| Mozilla Common Voice Scripted Speech 27.0 (opens in a new tab)Mozilla Foundation (Mozilla Data Collective) | Read speech: short clips of people reading given Bulgarian sentences aloud, with the sentence as transcript; 22 hours recorded, 18 hours validated. | CC0 1.0 | Allowed |
| FLEURS (opens in a new tab)Google | Read sentences from the FLoRes translation benchmark, recorded for 102 languages; training sets hold around 10 hours per language. | CC BY 4.0 | Allowed |
Common Voice and FLEURS are read, VoxPopuli lacks Bulgarian transcripts, and MATERIAL is paid-license phone conversation about 40% transcribed; ours is fully transcribed two-person conversation recorded in Bulgaria.
Licenses as published by each source, checked October 2026. Check the current terms before use.
FAQ
Frequently asked questions
What does the Bulgarian speech dataset include?
1,500 hours of spontaneous two-person conversations from 150 vetted native speakers recorded in Bulgaria, delivered as dual-channel audio with time-aligned transcripts, speaker metadata (age, gender, region, device), and consent documentation.
What makes Bulgarian hard for speech recognition?
Bulgarian dropped the Slavic noun cases but attaches a definite article to the end of words ("kniga", "knigata"). Stress is free and unstressed vowels get reduced, so pronunciation varies more than the Cyrillic spelling suggests.
Is there free Bulgarian speech data?
Yes, Mozilla Common Voice and FLEURS are free for commercial use. The MATERIAL Bulgarian-English Language Pack needs a paid license, and VoxPopuli has unclear terms. This dataset is spontaneous two-person conversation recorded in Bulgaria, with speaker metadata, one channel per speaker and signed commercial-use consent.
Can I use this Bulgarian dataset commercially?
Yes. The dataset is licensed for commercial use, including training ASR, TTS, and other voice-AI models. Every speaker has signed explicit commercial-use consent, and we provide the full consent chain for your legal review.
How much does the Bulgarian speech dataset cost?
Pricing is $85 per hour of audio. Volume discounts are available for full-dataset and multi-language purchases. Contact us for a quote.
Can I evaluate samples before buying?
Yes. Request samples with your company email, and we'll email you representative audio files and matching transcripts so your team can audit quality before committing.
$85/hr