South Asia · India
Punjabi conversational speech dataset
500 hours of spontaneous Punjabi conversations recorded in India across 50 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 | Punjabi |
|---|---|
| Recorded in | India |
| Speech type | Spontaneous two-person conversations |
| Volume | 500 hours |
| Speakers | 50 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 | $65 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 Punjabi speech:
- Automatic speech recognition (ASR / STT)
- Text-to-speech & voice synthesis prosody
- Voice agents and IVR for India
- 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 Punjabi. Read more in our guide to conversational vs. scripted speech data.
Free samples
Audit the Punjabi 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 Punjabi speech data vs. this dataset
| Source | What it is | License | Commercial training |
|---|---|---|---|
| IndicVoices (opens in a new tab)AI4Bharat | Read (8%), extempore (76%) and conversational (15%) speech from districts across India; includes Punjabi; gated download. | CC BY 4.0 | Allowed |
| Kathbath (opens in a new tab)AI4Bharat | IndicCorp sentences read aloud by contributors in Indian districts, collected on the Karya platform; includes Punjabi. | CC0 (dataset card text); Hugging Face metadata lists CC BY 4.0; AI4Bharat notes it does not own the IndicCorp source text | Allowed |
| Vaani (opens in a new tab)IISc Bangalore and ARTPARK, funded by Google | Spontaneous speech of people describing images, in many Indian languages including Punjabi; only a small share transcribed. | CC BY 4.0 | Allowed |
| 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 Punjabi sentences aloud, with the sentence as transcript; 4.4 hours recorded, 2.4 hours validated. | CC0 1.0 | Allowed |
The listed free Punjabi sets are read sentences, image-prompted speech or mostly extempore recordings; ours is spontaneous two-person conversation recorded in India, one channel per speaker, with commercial consent.
Licenses as published by each source, checked October 2026. Check the current terms before use.
FAQ
Frequently asked questions
What does the Punjabi speech dataset include?
500 hours of spontaneous two-person conversations from 50 vetted native speakers recorded in India, delivered as dual-channel audio with time-aligned transcripts, speaker metadata (age, gender, region, device), and consent documentation.
What makes Punjabi hard for speech recognition?
Punjabi is tonal, which is rare for an Indo-European language: tone changes the meaning of words that would otherwise sound the same. In India it's written in Gurmukhi, and speakers mix in Hindi and English freely.
Is there free Punjabi speech data?
Yes, IndicVoices, Kathbath, Vaani and Mozilla Common Voice are free for commercial use. This dataset is spontaneous two-person conversation recorded in India, with speaker metadata, one channel per speaker and signed commercial-use consent.
Can I use this Punjabi 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 Punjabi speech dataset cost?
Pricing is $65 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.
Related datasets
You might also need
Compare all 10 Indian language datasets by hours, speakers and price.