Imagicle provides AI-powered speech analytics through Imagicle Voice Analytics. The service offers multi-language speech-to-text transcription with automatic language detection, AI-based sentiment analysis that evaluates mood at overall call and per-segment level, and keyword/sentence tagging to classify topics. Dashboards, filters, and REST APIs expose sentiments, scores, transcripts, tags, and notes so organizations can monitor quality, detect recurring issues, and feed insights into external applications.

Imagicle Voice Analytics is a cloud-native, AI-powered add-on for Imagicle Call Recording that automatically transcribes recorded calls and analyzes them with advanced sentiment and keyword analysis. It turns recorded call audio into searchable text, dashboards, and alerts, so supervisors can quickly understand customer interactions, agent performance, and recurring issues instead of manually listening to hundreds of calls.

Voice Analytics applies a sophisticated speech recognition engine to convert the entire conversation into text, keeping local and remote parties separated. It can automatically detect the spoken language among 30+ supported languages, lets you search for keywords or full sentences, listen to specific sections, add notes, and export the full transcript. This saves time, makes it easier to find critical information, and supports faster, more accurate QA.

Yes. An AI engine analyzes intonation, cadence, and transcription content to determine the “mood” of the parties and label it as negative, neutral, or positive. Sentiment is available both at overall call level and for individual conversation segments, and a dedicated Sentiment Analysis Dashboard shows trends over time (overall, local party, remote party), making it easier to understand how calls went and how interactions are evolving.

In Voice Analytics you can add keyword or sentence tags to recordings (for example, product names, “refund”, “cancellation”, etc.). Once tagged, these keywords can be used as filters on the Recordings page to quickly retrieve calls that contain one or more tags and to group similar interactions, so supervisors can easily check which conversations match specific topics or processes.

Because every recording can be transcribed, scored for sentiment, tagged, and visualized in dashboards, supervisors can analyze all conversations instead of just a few samples. They can quickly see sentiment trends for users and remote parties, search inside specific calls, listen only to the most relevant sections, and use the insights (including mood evolution and tagged topics) to coach agents, improve scripts, and react precisely when issues emerge.

When a new recording is uploaded (either automatically from Imagicle Call Recording or manually as an MP3 file) Voice Analytics immediately starts transcription and sentiment analysis. Processing typically requires a few minutes depending mainly on call duration and the amount of audio being analyzed, after which transcripts, sentiment, and filters become available in the web portal.

Yes. If Screen Recording is enabled in Imagicle Call Recording, Voice Analytics lets you view captured screen and audio in a single player. From the recording detail you can play the audio waveform and, where available, open the “View screen recording” link to watch the synchronized screen capture, providing a complete view of what the agent said and did during the call.

The embedded speech recognition engine can automatically detect the language of a conversation out of over 30 supported languages. When uploading or transcribing recordings you can either manually select the language or let the engine detect it automatically, ensuring accurate transcription and sentiment analysis across multilingual environments. The detailed, up-to-date list of supported languages is available in the Voice Analytics Knowledge Base.

Voice Analytics runs within the Imagicle UCX Platform environment, where access is controlled via standard credentials or SSO (with optional multi-factor authentication) and role-based permissions determining who can see and manage which recordings. Each recording includes metadata with SHA256 checksums for audio and screen files to verify that content has not been altered. Recordings stored in Imagicle UCX Cloud Suite are kept in encrypted storage, and Imagicle Call Recording is designed to support strict regulations such as GDPR, HIPAA, PCI-DSS, and other compliance standards; Voice Analytics analyzes those compliant recordings on top of the same secure platform.

Yes. The Recordings page provides an advanced search bar and filters that let you: search by keyword or sentence across all transcribed recordings; filter by sentiment (any, negative, neutral, positive, or not available); and apply additional filters such as date range, call direction, duration, and tags. Within a single recording, you can also search inside the transcript to jump directly to specific words or phrases.

Yes. From the recording detail, the Download tab lets you download the audio file, any screen recording, and the full transcription. A dedicated “Transcription export” button is also available. For integrations, the Voice Analytics Recordings.v1 APIs expose recording metadata (including sentiment, tags, and transcript sentences) so that external systems and BI tools can retrieve and process this data programmatically.

Voice Analytics works on top of Imagicle Call Recording, making it available for the same environments as Call Recording: Microsoft Teams (hosted in Azure Private Cloud), Webex Calling (Dedicated Instance and Multi-Tenant), Cisco UCM, BroadWorks, Cisco CCME/CUBE, and any Media Gateway supporting the SIPREC protocol. While Call Recording can run in cloud, on-prem, or hybrid deployments, Voice Analytics is exclusively cloud.