Weekly Digest: What You Missed 🔥

Voice AI weekly digest

Top Updates 💪

• Voice technologies and AI: A bridge between humans and machines (Forbes)

• AWS & DXC collaborate to deliver voice translation to Amazon Connect (Amazon)

• OpenInfer raises $8M for AI inference at the edge (Venturebeat)

• Incept AI raises $3M in pre-seed funding (Finsmes)

• VoiceCare AI raises $3.85M in funding (Finsmes)

• Alexa is getting a major AI upgrade from Amazon. What we know so far (Cnet)

• Spotify partners with ElevenLabs for more AI-narrated audiobooks (TechCrunch)

• ComputerTalk’s innovative approach to contact center AI (Cxtoday)

• Deepgram hits milestone toward next-gen speech-to-speech (Morningstar)

• Conversational AI: Latest advances, future trends (IndiaTimes)

• Gladia: Breaking language barriers with AI speech recognition (Telecomreseller)

• RingCentral unveils AI Receptionist for customer calls (Telecomreseller)

• Healthcare conversational AI market: CureMetrix, Babylon, Google (Openpr)

• Aurora Mobile (JG) launches AI audio LLM for real-time voice interactions (Msn)

• RingCentral & BT launch RingCX for contact centers (Telecomreseller)

• Geely Auto, Stepfun open-source models for video, audio generation (Autonews)

• CEO Video: How Wordly advances live translation with AI (Meetings.skift)

• Actions technology is redefining the future of audio chips (Eetasia)

• Will podcasts become the key use case for AI dubbing? (Slator)

• Universal STT model leads in English, German, and Spanish (Assemblyai)

  

Voice AI Podcast 🎙️

In case you missed the latest episode of Voice AI Podcast…

---

Engineering Corner 😎

• This open TTS model needs just seconds of audio to clone your voice (Theregister)

• Step-Audio: Apache 2.0-licensed end-to-end voice model (Github)

• Speech recognition for different dialects and accents (Researchgate)

• The 10 best AI tools to convert text to audio (Marketing4ecommerce)

• Cascaded speech translation systems outperform end-to-end models (Slator)

• How to extract notes from videos using AI: Top tools compared (Techtricksworld)

• Microphone array geometry independent multi-talker distant ASR: NTT system for the DASR task of the CHiME-8 challenge (Arxiv)

• New AI ASR model cuts memory use by 80% while maintaining accuracy (Dev)

• Evaluating and selecting Wav2Vec 2.0 as Bajra’s core ASR solution (Blog.bajratechnologies)

• Improving diacritical Arabic speech recognition: Transformer-based models with transfer learning and hybrid data augmentation (Mdpi)

• Practical application of speech synthesis and model optimization in the intelligent voice assistant of HarmonyOS Next (Dev)

• End-to-end speech recognition with deep fusion: Leveraging external language models for low-resource scenarios (Mdpi)