Audio · Text / AI & Machine Learning
Speech ML | Keyword Spotting (Hard-to-Source Languages) | GDPR, CCPA Compliant | Native Speakers 180+ Countries
Advertised by providerLicensing rights not independently verifiedPrice on request
Overview
Speech ML | Keyword Spotting (Hard-to-Source Languages) | GDPR, CCPA Compliant | Native Speakers 180+ Countries is an audio and text dataset. Typical applications include speech recognition, speech synthesis (TTS), wake word & keyword spotting and 2 more. Relevant sectors: AI & Machine Learning, Education and Manufacturing & Industrial. The listing mentions transcription, timestamps and classification labels. Pricing is not published and is provided on request.
Primary applications
What the listing offers
- Listing mentions transcription, timestamps, classification labels.
Scope Short, prompted utterances and wake words such as “Go,” “Stop,” “Hey Siri”, and domain-specific command phrases. Each user contributes 150–200 samples, making this dataset ideal for wake-word detection, IoT devices, embedded systems, and intent classification. Available in any language or dialect, including rare and underrepresented varieties. Audio Specs • Sample rate: 16 kHz, 16-bit • File formats: WAV or MP3 • Configurable background noise levels and environments (quiet room, public space, in-vehicle, etc.) • Full control over speaker demographics, accents, and metadata granularity Volume Tens of thousands of utterances per project; scalable to millions of samples across diverse speakers and languages. Speakers Native speakers from our 1.2M+ user community in 180+ countries, with the ability to target specific demographics, regions, and rare dialects. Transcripts / Annotations • Keyword-level transcription and validation • Metadata tags (speaker ID, environment, device type, demographic details) • Optional timestamping for sequence training Why Us • Access to hard-to-source languages and accents for robust coverage • Fast, on-demand collection tailored to client needs • Rich metadata for training context-aware keyword models • Datasets collected in authentic real-world environments, ensuring production-level robustness Use Cases • Wake-word detection for virtual assistants (“Hey Siri,” “OK Google”) • IoT and smart device command recognition • Embedded keyword models
Related datasets
More in Video & Physical AI →Audio · TextAdvertised
Speech Data | OTS 100,000+ Hours of Free Speech Global + NICHE LANGUAGES | GDPR, CCPA Compliant | Native Speakers and Proprietary Data
Clean, high-quality unscripted recordings aligned with transcripts, built for training expressive and natural-sounding TTS systems.
- Language
- —
- Volume
- 100k h
- Domain
- AI & Machine Learning +2
- Use case
- Speech recognition
Price on request
More speech & audio datasets
All 1,727 speech & audio datasets →- Speech Recognition Data Collection Services | 100+ Languages Resources | Audio Data | Speech Data | Machine Learning (ML) Data
- Speech Recognition Dataset [Customer Calls] – Transcribed support conversations for training voice AI systems
- Speech Synthesis Data | 400 Hours | Text-to-Speech(TTS) Data | Audio Data | AI Training Data | Speech Data
- Speech Synthesis Data Collection Service | 50+ Languages Resources | Numerous Voice Sample | Audio Data | Text-to-Speech(TTS) Data
- Spontaneous Conversational Indonesian (Bahasa Indonesia) Speech Dataset — Consented, Transcribed
- Spontaneous conversational speech, healthcare and call-center contexts
- Spontaneous Conversational Sundanese Speech Dataset — Consented, Transcribed (West Java, Indonesia)