Top updates from 2026
August 2026
- soniox.com
Compare real-time speech translation APIs on your own audio
Soniox introduced a live demo and pricing calculator to compare its real-time speech translation API against competitors like OpenAI. The tool highlights Soniox's unified pricing model ($180 per 1,000…
- Featuresoniox.com
Compare text-to-speech APIson your own text
Soniox introduced a live comparison tool for text-to-speech APIs, allowing users to test Soniox, OpenAI, ElevenLabs, Google, Cartesia, and Azure on the same text in real time. The tool emphasizes Soni…
- soniox.com
Don't trust benchmarks.Test on your own audio.
Soniox introduced a live demo and pricing calculator to compare speech-to-text APIs side-by-side using real audio, emphasizing accuracy, cost, and production-ready features like diarization and multil…
- Featuresoniox.com
Data residency
Soniox launched a data residency feature allowing customers to select regions where their audio, transcripts, and metadata are processed and stored. System data (account metadata, usage stats, billing…
- Featuresoniox.com
Soniox vs Gemini 3.5 Live Translate
Soniox published a side-by-side comparison of its real-time translation capabilities against Google's Gemini 3.5 Live Translate, emphasizing cost savings (92% cheaper at 1,000 hours/month), any-to-any…
- Launchsoniox.com
Soniox vs OpenAI for real-time speech translation
Soniox introduced a real-time speech translation comparison tool pitting its service against OpenAI’s GPT Realtime Translate, highlighting cost savings (91% cheaper at 1,000 hours/month) and broader l…
- soniox.com
Language identification
Soniox’s Speech-to-Text AI now automatically identifies spoken languages in audio streams, including mixed-language conversations, without requiring users to pre-specify languages. The feature tags ea…
- Featuresoniox.com
Speaker diarization
Soniox introduced speaker diarization in its Speech-to-Text AI, enabling automatic detection and separation of speakers in multi-speaker audio streams. The feature generates speaker-labeled transcript…
- Featuresoniox.com
Endpoint detection
Soniox introduced semantic endpoint detection for real-time speech-to-text, using pauses, intonation, and context to finalize utterances earlier than silence-based methods. The feature reduces latency…
- Launchsoniox.com
MeetSonioxText-to-Speechv2
Soniox released TTS v2, introducing programmable audio tags for expressive control, support for over 60 languages with natural mixing, high-quality voice cloning from seconds of audio, and low-latency…
July 2026
- Launchsoniox.com
Introducing Soniox Compare
Soniox introduced Soniox Compare, an open-source suite of tools enabling side-by-side testing of voice AI providers using real-world audio, text, and languages. Users can compare speech-to-text, text-…
- Launchsoniox.com
Introducing Soniox Voice Cloning
Soniox introduced Voice Cloning, a technology that creates high-fidelity digital replicas of voices from short audio samples, capturing unique vocal traits like tone, rhythm, and emotion. The system s…
June 2026
- Launchsoniox.com
Soniox v5 Real-Time: Turning live conversations into structured intelligence
Soniox introduced Soniox v5 Real-Time, a streaming speech AI system that converts live audio into structured, machine-readable data with native speaker separation, language identification, and real-ti…
- Launchsoniox.com
Soniox v5 Async: Turning real-world speech into structured data
Soniox introduced Soniox v5 Async, a major upgrade to its speech-to-text AI designed to convert noisy, multilingual, and overlapping real-world audio into structured, machine-readable data. Key improv…
- Partnershipsoniox.com
Partnership between Tencent Cloud and Soniox
Soniox announced a strategic partnership with Tencent Cloud, integrating its Speech-to-Text API directly into the Tencent RTC console. This enables developers to build real-time, multilingual voice AI…
May 2026
- Contentsoniox.com
Everything you dowith voiceinFinnish
Soniox highlights the capabilities of its voice technology specifically for the Finnish language, emphasizing features and use cases available in their app.
- soniox.com
One platform for multilingual voice AI
Soniox introduced a single API platform unifying speech-to-text, text-to-speech, and translation across 60+ languages with native-speaker accuracy. The platform targets real-time applications, support…
- Launchsoniox.com
The world’s most accurate speech-to-text
Soniox unveiled a new speech-to-text platform delivering native-speaker accuracy across 60+ languages, including robust handling of multilingual speech, precise alphanumeric recognition, low-latency s…
- Contentsoniox.com
Speech-to-text API comparison
Soniox launched an open-source comparison tool testing real-time transcription accuracy, latency, and feature coverage across major STT providers using identical audio inputs. The tool highlights Soni…
- Featuresoniox.com
Power wearable apps with fast, accurate speech recognition
Soniox updated its speech-to-text API to support real-time transcription and translation for wearables like smartwatches, glasses, and earbuds. The API processes audio locally with no storage, support…
Top 20 shown. Click a month above for the full archive of Soniox.
Track Soniox on autopilot
- · Weekly AI brief: narrative summary of what shipped, every Monday 9 AM
- · Email or Slack alerts, or chat with the archive in your dashboard
- · Add Soniox + up to 2 more competitors free, no credit card