Why CallMeter

The VoIP testing market has a gap that has persisted for over a decade. On one side, free open-source tools like SIPp test SIP signaling but cannot tell you whether calls actually sound good. On the other side, enterprise testing platforms charge $50,000 to $500,000+ per year, require lengthy procurement cycles, and still lack the metric depth, video instrumentation, and flexibility that modern telecom teams need.

Neither side serves the teams that need VoIP quality testing the most: the carrier engineer validating a new SIP trunk before go-live, the IT director migrating 50 offices to a new UCaaS platform, the DevOps team that wants to gate deployments on voice quality, or the MSP trying to prove SLA compliance to dozens of clients.

CallMeter was built to close that gap. It is a purpose-built platform for SIP & WebRTC stress testing and quality monitoring that combines real media generation, 150+ per-endpoint metrics, continuous monitoring, and self-service pricing into a single product that any telecom professional can use without a protocol engineering degree or a six-figure budget.

---

The Market Problem

Telecom engineers today face a set of frustrating trade-offs when choosing how to test voice and video infrastructure.

Free Tools: Signaling Without Quality

The most popular open-source SIP testing tool, SIPp, was designed when SIP signaling was the hard problem. It generates SIP messages at high throughput and validates that your server handles INVITE, REGISTER, BYE, and other methods correctly. This remains useful for capacity planning and SIP proxy regression testing.

But SIPp does not generate real RTP media. It replays pre-recorded PCAP files rather than negotiating codecs dynamically. It cannot measure MOS, jitter, packet loss, round-trip time, or any other quality metric. It cannot test video at all. It tests whether calls connect, not whether they sound good. In modern deployments where quality is the differentiator, signaling-only testing misses the most important question.

Other free tools share similar limitations. baresip is a softphone, not a testing platform — it makes calls but does not instrument them. Teams that try to build testing stacks around baresip quickly discover that extracting deep quality metrics from a tool designed for phone calls requires custom C development, fragile scripting, and ongoing maintenance. Commercial platforms that wrap baresip as their media engine inherit the same observability ceiling. No combination of open-source tools offers dashboards, continuous monitoring, APIs, or team collaboration out of the box.

Enterprise Platforms: Comprehensive but Inaccessible

At the enterprise end, legacy testing platforms offer end-to-end VoIP quality assurance. However, they come with significant barriers:

• Cost: $50,000 to $500,000+ per year. This prices out the vast majority of telecom teams.
• Procurement: Enterprise sales cycles take weeks to months. You cannot start testing today.
• Focus: Many are built for CCaaS quality assurance or carrier-grade network monitoring, not general-purpose SIP infrastructure testing.
• Complexity: Legacy deployment models with proprietary hardware, thick clients, or ecosystem lock-in.
• Vendor lock-in: Proprietary data formats, long-term contracts, and switching costs that compound over time.

Manual Testing: The Default Fallback

When tools are either too limited or too expensive, teams fall back to manual testing: an engineer dials a test number, listens to the audio, and declares it "sounds fine." This is unscalable, unquantifiable, unreproducible, and catches problems only after users are already affected.

---

10 Things That Make CallMeter Different

CallMeter was designed from the ground up to address the limitations of every existing approach. Here are the ten differentiators that set it apart.

1. Real Media Processing

CallMeter does not simulate media or replay packet captures. Every test endpoint performs actual codec negotiation via SDP offer/answer and encodes and decodes real audio and video using production codecs. The media path is fully exercised: RTP packetization, jitter buffering, codec processing, RTCP feedback loops, and FEC recovery all happen exactly as they would in a production call.

This means your test results reflect what real users will experience. If a SIP trunk introduces transcoding artifacts, if a session border controller adds unexpected jitter, or if a codec mismatch causes fallback to a lower-quality codec, CallMeter detects it because the media is real.

2. 150+ Real-Time Metrics Per Endpoint

Every endpoint collects over 150 measurements per second across eight categories:

Every metric is separated by direction (send and receive), giving you visibility into both sides of the media path. This directional separation is critical for diagnosing asymmetric quality issues, such as one-way audio or degradation that only affects the receive path, that aggregate metrics completely obscure.

This is the deepest per-call instrumentation available in any SIP testing platform, open-source or commercial. Most enterprise platforms document 10 to 50 metrics. No open-source tool measures any of these.

3. Audio AND Video Testing

Video calling is standard in enterprise communications, yet most SIP testing tools only support audio or ignore video quality entirely. CallMeter supports seven codecs across both modalities:

You can test voice-only SIP trunks with PCMA, wideband enterprise calls with G.722, adaptive WebRTC sessions with Opus, and video conferencing infrastructure with H.264 or VP8, all from the same platform with the same depth of quality instrumentation.

Custom media file uploads allow you to send specific audio or video content during tests, making scenarios more realistic and enabling targeted quality assessments.

4. Cloud Workers + Self-Hosted Workers

CallMeter offers two deployment models for test endpoints:

Cloud workers are hosted and managed by CallMeter in multiple regions. They are available immediately, require no infrastructure, and are ideal for testing internet-accessible SIP infrastructure. Select a region near your users or your SIP servers and start testing in minutes.

Self-hosted workers are lightweight Docker containers that you deploy on your own infrastructure. They connect outbound to CallMeter (no inbound firewall rules required) and enable testing from inside private networks, behind firewalls, in specific data centers, or anywhere that matters for your quality assessment.

This hybrid model gives you the reach of a cloud platform with the access of on-premise tooling. No other testing platform offers both with the same simplicity.

5. Modern Web Dashboard

CallMeter is built for teams, not just SIP protocol engineers. Anyone on the team, whether they are a network engineer, an IT director, a QA lead, or an operations manager, can create tests, view real-time results, analyze historical data, and share findings through a modern web interface.

Key dashboard capabilities:

• Real-time test monitoring with per-second metric updates during test execution
• Per-endpoint drill-down to isolate quality issues to specific call legs
• Time-series charts for every metric with send/receive direction filtering
• Historical comparison across test runs to detect regressions
• Multi-tenant organization with five-level RBAC (Owner, Admin, Editor, Operator, Viewer)
• Per-project scoping so teams see only the data relevant to their work

No CLI required. No XML authoring. No log file parsing. The dashboard makes 150+ metrics accessible to everyone on the team.

6. Continuous Monitoring with Probes

Stress tests answer "can my infrastructure handle load?" Probes answer the equally important question: "is my infrastructure healthy right now?"

CallMeter probes run scheduled SIP calls at configurable intervals (every 5, 15, 30, or 60 minutes), measure quality metrics on every execution, and evaluate results against thresholds you define. When quality degrades, probes transition through health states (Healthy, Degraded, Unhealthy) and trigger webhook notifications to your alerting system.

Probes power public status pages that you can share with customers, partners, and regulators. These dashboards display real-time and historical quality data, providing transparent, always-on proof that your SIP infrastructure meets its commitments.

This continuous monitoring catches problems between tests, before users report them, and creates a permanent quality audit trail.

7. Group-Based Testing

Real-world call scenarios are rarely uniform. A production environment might have PSTN trunks using PCMA, wideband enterprise calls on G.722, and WebRTC clients using Opus, all running simultaneously through the same infrastructure.

CallMeter's group-based testing lets you model these complex scenarios in a single test:

• Multiple groups within one test, each with its own registrar, codec, media files, and worker assignment
• Independent configuration per group: different SIP accounts, transport protocols, call durations, and endpoint counts
• Cross-group targeting where callers in one group can call endpoints in another group, simulating real multi-party call flows
• Per-group and per-endpoint metrics so you can compare quality across different configurations in the same test run

This replaces the custom scripts and multi-tool orchestration that teams typically build around tools like SIPp to simulate realistic traffic patterns.

8. Historical Trending and Analytics

Every metric from every call in every test run is stored and queryable. This is not just last week's data; it is every measurement from every test you have ever run on the platform.

This persistent storage enables:

• Regression detection: Compare metrics across test runs to spot quality degradation after infrastructure changes
• Capacity planning: Trend quality metrics against endpoint count to find the load level where quality breaks down
• SLA reporting: Export historical probe data to prove compliance over any time period
• Root cause analysis: Drill into specific time ranges, endpoints, and metrics to understand exactly when and where quality degraded

Unlike open-source tools that produce ephemeral output lost after each run, CallMeter builds a comprehensive quality history that becomes more valuable over time.

9. Transparent Pricing

CallMeter is self-service from day one. Create an account, choose a plan, and start testing. No sales calls. No procurement cycles. No six-figure contracts. No surprises.

• 7-day trial on all paid plans so you can evaluate before committing
• Transparent pricing so you know what you pay before you commit
• Subscription model with clear credit-based usage so costs are predictable
• Upgrade or downgrade at any time as your needs change
• No minimum contract on self-service plans

For teams that need higher volume or dedicated infrastructure, enterprise options are available through the sales team. But the majority of teams can self-serve from signup to production monitoring without ever talking to a salesperson.

10. API and CI/CD Integration

CallMeter exposes a REST API for programmatic test execution, result retrieval, and quality gate integration. This enables:

• CI/CD quality gates: Trigger tests on every deployment, evaluate results against thresholds, and block releases that degrade voice quality
• Automated regression testing: Run standardized test suites against staging environments before promoting to production
• Custom dashboards: Fetch metric data and integrate it into your existing monitoring stack
• Scheduled workflows: Trigger tests from external schedulers, orchestration tools, or event-driven pipelines

A typical CI/CD integration flow:

1. Deploy a SIP infrastructure change to staging
2. API call triggers a CallMeter test against the staging environment
3. Pipeline waits for test completion
4. Pipeline fetches results and checks: MOS above 3.8? Packet loss below 1%? PDD below 500ms?
5. Pass: promote to production. Fail: block and notify.

This transforms VoIP quality from a manual post-deployment check into an automated pre-deployment gate.

---

Why CallMeter Built Its Own Media Engine

Most testing platforms take a shortcut: they wrap an existing softphone (typically baresip) and add a web interface on top. This is faster to build but creates a permanent architectural ceiling. The softphone's internals are a black box — you can measure what it exposes, but you cannot measure what it hides. And softphones are designed to hide errors from users, not reveal them as data.

CallMeter took a different path. Every component — from SIP signaling to codec processing to metric collection — is purpose-built for testing and instrumentation. This is not about wrapping a better softphone. It is about controlling every layer of the media pipeline so that nothing is hidden.

What this architecture enables:

• 150+ metrics per second, streamed live. Not collected after the call ends — streamed to the dashboard in real time as the call happens. You watch quality evolve second by second, not in a post-mortem report.
• Video as a first-class citizen. H.264, VP8, and VP9 get the same depth of instrumentation as audio: freeze detection, freeze duration, resolution tracking, keyframe analytics, FPS monitoring, and per-frame decode quality. Video is not an afterthought — it is instrumented at the same depth as audio.
• Codec-level observability. Every stage of the codec pipeline is instrumented: packet loss concealment events, voice activity detection, comfort noise generation, Opus bandwidth adaptation, decoder errors, and more. Softphones hide these events to improve user experience. CallMeter reveals them as data.
• Dual metric collectors (send and receive) that validate SSRC ownership per RFC 3550 — measuring both sides of every call independently to diagnose asymmetric quality issues.
• Zero-loss metric queue that guarantees no measurement is dropped under any load condition, backed by mutex-protected queues.
• Network impairment injection — inject controlled packet loss, jitter, latency, and bandwidth constraints into the media path to test how infrastructure handles degraded conditions. This requires deep integration with the media pipeline that softphone-based platforms cannot achieve.
• Custom pass/fail thresholds on any metric. Set thresholds on freeze count, jitter buffer underruns, audio level, resolution drops, PLC events, or any combination of 150+ measurements. You define what a successful call means — not the platform.
• Continuous monitoring with probes that run scheduled calls, evaluate quality against your custom thresholds, transition health states (Healthy, Degraded, Unhealthy), fire webhooks, and power public status pages — all driven by the same deep instrumentation.

This is why CallMeter measures 150+ metrics per endpoint where softphone-based platforms measure approximately 15 — and why it offers capabilities like network impairment injection, custom threshold evaluation, and live metric streaming that are architecturally impossible on a softphone foundation.

---

Who Uses CallMeter

CallMeter serves telecom professionals across four primary verticals:

For detailed use cases with pain points, solutions, and key features, see Use Cases.

---

What to Read Next