Twilio
Twilio's voice AI stops cutting callers off mid-sentence, using its own turn-taking model.
Comparison · Comms
Speechmatics vs Rspamd
A side-by-side editorial comparison of Speechmatics and Rspamd — release velocity, themes, recent moves, and the top alternatives to consider.
Speechmatics vs Rspamd: at a glance
| Feature | Speechmatics | Rspamd |
|---|---|---|
| Sector | Comms | Comms |
| Velocity score | 5.0 | 5.0 |
| Sparks · 30d | 0 | 0 |
| Top themes | speech-to-text, voice agents, multilingual, medical | spam filtering, email authentication, input hardening, attachment extraction |
| Last editorial update | 6mo ago | 2h ago |
| Website | — | Visit → |
What is Speechmatics?
Speechmatics rolls its Enhanced English model across the stack, citing 89% WER gains on spellouts.
Speechmatics is a speech-recognition platform whose last quarter has been a coordinated rollout of its Enhanced Operating Point English model from containers through realtime and batch SaaS. The accuracy story is unusually concrete: 69% relative WER improvement on numbers, 89% on spellouts, 42% on mixed alphanumerics. Alongside the model work, the platform is adding voice-agent ergonomics — End of Utterance detection, prefer_current_speaker, speaker sensitivity — and broadening bilingual coverage.
What is Rspamd?
Rspamd 4.3 fixes SPF evaluation order and rejects weak DKIM keys, with results that change.
Rspamd 4.3.0 on November 13 is a correctness release with a long incompatible-changes list. SPF records are now evaluated in term order, closing a case where an explicit -ip4 deny could be overridden by a broader allow. DKIM rejects RSA keys under 1024 bits and signatures with duplicate tags, DMARC no longer drops a policy over an unknown tag, URL mismatches are scored apart from phishing, and PDF text extraction and archive unpacking change their output. Critical fixes cover a header line that could abort a worker from one unauthenticated HTTP request.
Speechmatics vs Rspamd: editorial side-by-side
Speechmatics
5.0COMMS
Speechmatics rolls its Enhanced English model across the stack, citing 89% WER gains on spellouts.
◆ Current state
Speechmatics is a speech-recognition platform whose last quarter has been a coordinated rollout of its Enhanced Operating Point English model from containers through realtime and batch SaaS. The accuracy story is unusually concrete: 69% relative WER improvement on numbers, 89% on spellouts, 42% on mixed alphanumerics. Alongside the model work, the platform is adding voice-agent ergonomics — End of Utterance detection, prefer_current_speaker, speaker sensitivity — and broadening bilingual coverage.
◆ Where it's heading
Two threads are running in parallel. Vertical depth: domain-specific models starting with medical, plus a growing list of bilingual code-switching pairs (Tagalog, Malay/English, Tamil/English, Mandarin/English, Arabic/English). Horizontal coverage: each model lands in containers first, then realtime SaaS, then batch SaaS, then appliance — containers function as the proving ground and SaaS as the broad rollout vehicle. The release notes also hint at voice agents being the primary use case Speechmatics is optimising for.
◆ Prediction
Expect more vertical-domain Enhanced models beyond medical (legal and finance are the obvious next targets) and a tighter packaging of the voice-agent primitives — End of Utterance, current-speaker locking, low-latency operating points — into something explicitly marketed as a voice-agent SDK or recipe.
Rspamd
5.0COMMS
Rspamd 4.3 fixes SPF evaluation order and rejects weak DKIM keys, with results that change.
◆ Current state
Rspamd 4.3.0 on November 13 is a correctness release with a long incompatible-changes list. SPF records are now evaluated in term order, closing a case where an explicit -ip4 deny could be overridden by a broader allow. DKIM rejects RSA keys under 1024 bits and signatures with duplicate tags, DMARC no longer drops a policy over an unknown tag, URL mismatches are scored apart from phishing, and PDF text extraction and archive unpacking change their output. Critical fixes cover a header line that could abort a worker from one unauthenticated HTTP request.
◆ Where it's heading
Since 4.1, every release has hardened parsers and closed memory-safety and trust gaps; 4.3 applies the same pass to the authentication standards themselves (SPF, DKIM, DMARC, ARC) and to content extraction. Operators should expect scores and fuzzy hashes to shift after upgrading, because the project now favours spec-correct results over backward-compatible ones.
◆ Prediction
A 4.3.x point release fixing regressions from these behaviour changes is the likely next step, as 4.2.1 followed 4.2.0 within two weeks.
Alternatives to Speechmatics and Rspamd
Other Comms products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either Speechmatics or Rspamd.
Salesmsg
Salesmsg adds call-back requests and a one-click Emergency Stop for all outbound texts.
Element
Element 1.12.35 follows the call picture-in-picture release with one desktop storage fix.
Krisp
Krisp moves call transcripts and live captions to cloud transcription in KrispCX 2.84.4.
Misskey
Misskey 2026.11 ships: a leaner start script, and Node.js 22 no longer boots.
SlickText
SlickText's feed is all SMS buying guides; no product release has surfaced since RCS.
See all Speechmatics alternatives → · See all Rspamd alternatives →
Recent activity from Speechmatics and Rspamd
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 21h agoRspamd4.3.0: SPF order fix, stricter DKIM, new truncation symbols
- 24d agoRspamd4.2.3: ARC trust fix, merged RBL queries, browser-style HTML parsing
- 1mo agoRspamd4.2.2: crypto wallet detection and changed MIME digests
- 1mo agoRspamd4.2.1: critical HTTP and memory-safety fixes
- 1mo agoRspamd4.2.0: public suffix lookup and attachment content extraction
- 3mo agoRspamdCloses an arbitrary file read reachable by any TCP client
- 6mo agoSpeechmatics15.7.0 - Containers ⚡
- 6mo agoSpeechmatics2026.04.23 Realtime SaaS
- 7mo agoSpeechmatics2026.04.16 Batch SaaS
- 8mo agoSpeechmatics2026.03.12 - Realtime SaaS
- 8mo agoSpeechmatics2026.03.09 - Batch SaaS
- 8mo agoSpeechmatics1.2.0 Realtime Kubernetes — Redis repopulation and CVE patches
Frequently asked questions
What is the difference between Speechmatics and Rspamd?
They serve adjacent needs but don't currently overlap on shipped themes. Speechmatics and Rspamd are shipping at a similar cadence (velocity 5.0 vs 5.0, both within Sparkpulse's "active" band). See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Is Speechmatics better than Rspamd?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. Speechmatics and Rspamd are shipping at a similar cadence (velocity 5.0 vs 5.0, both within Sparkpulse's "active" band). For your specific use case, the alternatives sections above list other Comms products to evaluate alongside.
What are the best alternatives to Speechmatics?
Top Speechmatics alternatives in Comms are ranked by recent ship velocity. Browse the "Speechmatics alternatives" section above for the current picks, or visit /alternatives/speechmatics for the full list with editorial commentary on each.
What are the best alternatives to Rspamd?
Top Rspamd alternatives in Comms are ranked by recent ship velocity. Browse the "Rspamd alternatives" section above for the current picks, or visit /alternatives/rspamd for the full list with editorial commentary on each.