Speechmatics

What is Speechmatics

Speechmatics is a foundational speech technology platform that combines accurate speech recognition and AI capabilities. It offers real-time transcription, translation, and understanding in 50 languages, with high accuracy even in challenging environments. It has partnerships with companies like Red Bee Media and Limecraft, and its speech intelligence technology allows for the analysis of audio events beyond just words. It provides transformative advantages with its real-time speech technology and offers various features and deployments for different use cases like contact centers, media captioning, and edtech.

Key Features of Speechmatics

- Multi-Language Support: Supports transcription, translation, and understanding in 50 languages, covering over half the world's population and helping businesses expand globally.

- Real-Time Capabilities: Provides precise, low-latency transcription and translation in real-time, delivered before the media even ends.

- Unmatched Accuracy: Delivers high performance across diverse voices, even in challenging and noisy environments, ensuring reliable outputs.

- Single API Integration: Combines accurate speech recognition with the latest AI and LLM technology through a single API.

- Speech Intelligence: Introduces features like Audio Events to enhance content analysis, media analytics, and accessibility efforts.

- Case Study Success: Partners like Red Bee Media and Limecraft highlight the tool’s superior quality metrics, such as word error rate, speaker segmentation, and punctuation.

- Developer and Business-Friendly: Facilitates easier extraction of insights from online meetings through partnerships, such as with Recall.ai for real-time transcription.

- Comprehensive Features: Offers a range of functionalities including transcription, summarization, chapters, and more for diverse use cases like Contact Center Solutions, Media & Event Captioning, and Meeting Platforms.

Pricing

Free Plan:

• Cost: $0/month
• Monthly Hours: 8 hours free (4 hours batch + 4 hours real-time)
• Features:
  • 50 languages supported
  • Standard or Enhanced accuracy
  • Industry-leading accent coverage
  • Real-time latency <1s
  • Language identification
  • Speaker diarization (Real-time and Files)
  • Custom dictionary
  • Precise timestamps
  • Advanced punctuation and casing
  • Numeral formatting
  • Profanity and disfluency detection
  • Multi-channel files supported
  • Export SRT captions
  • Audio events
  • Speech Capabilities like Translation, Summaries, Chapters, Sentiment, Topics
  • SaaS deployment
  • Rate limits: 2 concurrent real-time sessions

Pay As You Grow:

• Cost: Starting at $0.30 per hour
• Batch transcription (Pre-recorded):
  • $0.30/hr for Lite Mode Standard accuracy
  • $0.80/hr for Standard accuracy
  • $1.04/hr for Enhanced accuracy
• Real-time transcription (Live stream): $1.35/hr
• Speech Capability Bolt Ons:
  • $0.65/hr for Translation
  • $0.12/hr for Summaries
  • $0.40/hr for Chapters
  • $0.15/hr for Sentiment
  • $0.20/hr for Topics
• Features:
  • 10 concurrent real-time sessions
  • 10 batch jobs per second
  • Online email support included
  • SaaS deployment

• Enterprise Plan:
• Cost: Custom pricing
• Features:
  • Volume savings available
  • Unlimited scale with flexible deployments
  • Audio alignment
  • Multiple deployment options: SaaS, Private Cloud, Container, Virtual Appliance, On-Device
  • GPU and CPU based models available
  • Choice of cloud region: US, EU or Australia
  • No rate limits
  • Prioritized enterprise support
  • Eligible for Early Access Features
  • Dedicated Customer Success and Sales Engineer
  • Custom models and Language model adaptation
  • Customer community access