Deepgram Reviews

Deepgram.com offers a product for transcription that is powered by advanced AI. The key features and benefits include:

• Unmatched performance and value: The next-gen speech-to-text models surpass all competitors in speed, accuracy, and cost. They are trained to handle background noise, multiple speakers, and cross-talk during podcasts and recorded or live video streaming to provide accurate, readable captioning at a competitive price.

• Smart transcript formatting: Transcripts come with built-in speaker labels (diarization) and word timings, enhancing readability and streamlining workflows. The formatting includes automatic punctuation and paragraphs, contextualized entities, alphanumerics, and more.

• Real-time results: Low-latency streaming transcription, long audio file handling, and up to 20x faster caption creation of pre-recorded audio content than alternatives.

• Language AI models: These models can create accurate summaries and identify speaker sentiment, topics, and intent to facilitate derivative content creation and analytics.

• Rich content captioning: Transcripts are easy to read and super accurate, making them suitable for accessibility, usability, or compliance.

• SEO and audience expansion: Adding transcripts enables search engines to crawl and index your content, expand your audience, and increase engagement.

• Content moderation: Quickly flag sensitive content like profanity and hate speech to ensure audience and brand safety.

• Searchability and user experience: Create rich summaries and searchable transcripts that enable your audience to quickly jump to precise moments in specific podcasts and videos of interest.

• Streamline workflows: Automate subtitling tasks with the most accurate transcripts in the market that include speaker labels and smart formatting for free.

• Content analytics: Use Language AI to analyze sentiment and topics of your programming and see how it correlates with audience engagement.

• Support for Over 36 Languages: Catering to a global audience by supporting multiple languages and dialects.

• Various Input Methods: Flexibility to transcribe live speech, audio files, or even YouTube videos.

• High Accuracy: Leveraging advanced AI to ensure that your transcriptions are as accurate as possible.

Deepgram.com offers several pricing plans for its transcription product:

• Pay As You Go: This plan is free with $200 of credit. Then it's pay-as-you-go with no minimums, no expiration, and no credit card required. It allows up to 100 concurrent requests for Deepgram speech-to-text models and 5 concurrent requests for Deepgram Whisper Cloud.

• Growth: This plan costs between $4k-10k per year and offers up to 20% savings with pre-paid credits for the year. Credits are redeemed against actual usage.

• Enterprise: This plan is exclusively for businesses with large volumes, data or deployment requirements, or support needs. It offers access to all endpoints and public models with the best discounts.

For Speech to Text services, the pricing is as follows:

• Nova-2 (pre-recorded): $0.0043/min
• Nova-2 (streaming): $0.0059/min
• Whisper Cloud (pre-recorded): $0.0048/min
• Custom Redaction: $0.0020/min

For Text to Speech services, the pricing is as follows:

• Aura: $0.0150/1k characters

For Audio Intelligence services, the pricing is as follows:

• Summarization: $0.0003/1k input tokens - $0.0006/1k output tokens

Please note that when using the multichannel feature, each channel is transcribed and billed separately. The total cost when using multichannel is the single-channel cost multiplied by the number of channels.

Deepgram.com offers a variety of payment methods and plans.

• Pay As You Go: This plan is free to start with, and provides $200 of credit. After the credit is used, customers pay for what they use without any minimum spend or expiration date. No credit card is required to start using the service.
• Growth Plan: This plan costs between $4k-10k per year. Customers can save up to 20% with pre-paid credits for the year. Credits are redeemed against actual usage.
• Enterprise Plan: This plan is for businesses with large volumes, data or deployment requirements, or support needs. It offers access to all endpoints and public models with the best discounts.

The costs for various services are as follows:

• Speech to Text: Costs range from $0.0043/min to $0.0145/min depending on the model used.
• Text to Speech: Costs $0.0150/1k characters.
• Audio Intelligence: Costs range from $0.0003/1k input tokens to $0.0006/1k output tokens.

Please note that this information is based on user reviews and may vary based on individual experiences.