About OTO

DeepTone Insights Anywhere is a web-based speech analytics SDK designed to help businesses detect speech, gender and language to gain insights into various metrics in human speech. Key features include audio capture, voice recognition, concatenated speech, sentiment analysis, call recording and data security. Teams using DeepTone Insights Anywhere can detect negative interactions and classify emotions as happy, tired, irritated and neutral to adjust empathy based on the responses, which help to prioritize mental well-being and promoting work-life balance. The arousal model allows contact centers to detect voice energy to facilitate agent engagement and real-time coaching. DeepTone Insights Anywhere enables gaming communities to use the gender model to identi...

OTO Pricing

OTO.ai is available on pay-as-you-go and monthly subscription plans, as follows: Starter (pay-as-you-go): $0.60 / processed hour Pro: $ 100 / month or $0.20 / processed hour

Starting price: 

$100.00 per month

Free trial: 


Free version: 

Not Available


Image 1 of 1

DeepTone Insights Anywhere voice markers

OTO User Reviews


Showing 1 review


Verified reviewer

Company size: 1 employee

Industry: Information Technology and Services

Review Source: Capterra

This reviewer was invited by the software vendor to submit an honest review.


out of 5


out of 5

January 2021

Perfect & Easy to use. - Strongly recommended.


Being a student, researching in a similar field of Speech Emotion Recognition/ Behavioural AI, I am really impressed with the product. I tried some labeled use-cases with DeepTone to find out results and compared them with original labels. Here are the things I liked the most about DeepTone. 1. The user interface. I found it really very interactive and easy to use. It raises your interest in the product. 2. Classification of Speech, Music, or Silence works completely fine. The voice model provides the results with high accuracy. 3. Classification of the voice of male or female provides accurate results. 4. The mapping of the energy of the speaker always shows correct results. I feel it gives more understanding of recognizing user behavior or behavior. 5. The best part is emotion recognition. I found it really amazing, which classifies the speech according to user behavior i.e. Happy, Irritated, Neutral, or Tired. I found emotion recognition of the DeepTone voice model works with great precision. Overall I think, this product can be the best option to find out customer behavior and it can be a really useful tool for reaching out to target customers. I surely recommend this to fellow businesses/customers.


Only one thing concerning I realized, sometimes the DeepTone model gives confusing results in emotions (Specifically Irritated voice and Happy voice) with high-energy tones. I think the doors are open for improvement in this type of scenario and it can also be an area of research. Overall I am really happy with the product.