NVIDIA Broadcast, the free RTX app streamers use to clean up their mic and webcam live, can now fix recordings after the fact. Version 2.3 adds a File Processor Mode for MP4, M4A, MP3 and WAV files.
OpenAI shuts down tts-1, tts-1-hd and gpt-4o-mini-tts on January 6, 2027. The replacement, gpt-realtime-2.1-mini, cannot serve the speech endpoint, returns no MP3 and drops onyx, nova and fable. We priced the switch, audited 9 creator tools and tested a free local drop-in.
DeepFilterNet3 lifted speech quality on every noise we tested, from keyboard clatter to mains hum, on one CPU core. At its default setting it also removes a music bed almost completely, and the RNNoise inside OBS is the pre-2024 model.
Hugging Face's new Open TTS Leaderboard puts the 82M-parameter Kokoro first in English. We read every model's license: in the multilingual and voice cloning views the #1 and #2 are non-commercial, and the first model you can ship is Qwen3-TTS at #3.
Fermion Research's Phonon-2, a 164 MB open English speech model released 29 September, transcribed an hour of audio in 39.3 seconds on our 6-thread Linux server, about 15x faster than Whisper large-v3-turbo, at level accuracy. Its hotwords fix for names only works on Macs.
ElevenLabs' Eleven v4 took first place on the Artificial Analysis Speech Arena at 1319 Elo on launch day. It costs $0.022 per 1,000 characters until 12 October, then $0.08: here is who should switch now and who should wait.
Most AI video clips come out silent. Here is how to add synced footsteps, ambience and impacts in about thirty minutes, and why the licence decides the tool before quality does.
Google's new Flash TTS and Flash-Lite TTS lead Hume AI's Voice Design Benchmark. The launch price is promotional, and every tier doubles on 1 January 2027.
Alibaba sold Qwen-Audio-3.1 with 70 to 95% price cuts. The biggest verifiable cut is on the older 3.0 Realtime Flash, whose audio output fell from $15 to $1.87 per million tokens.
You can rescue a badly recorded interview in about fifteen minutes using free tools, and the hard part is not which tool you pick. It is the order you run them in. AI speech enhancers are destructive, and stacking two of them is how a usable take becomes a silent waveform.
Alibaba released Qwen3.8-LiveTranslate on 19 September 2026, cutting average lag from 2.8 to 2.3 seconds across 60 languages. It charges more than twice Gemini's per-token price for audio and still costs about 30% less per hour. The number that decides your bill is not in the headline.
SpaceXAI shipped Grok Voice Transcribe 2.0 with two claims stacked on top of each other: twice as accurate as version 1.0, and first among 32 streaming models. Both are defensible. Neither is the reason the release matters.
Alibaba's new omni model takes text, images, audio and video at $0.15 per million input tokens, exactly a fifth of Gemini 3.8 Flash. The number that decides your bill is not on the pricing page.
Wispr published Canto on 17 September 2026 with a 3.4% word error rate on real dictations, the lowest of six models. Its own charts also show 21.4% on short dictations, where three models tie exactly.
Nari Labs says it leads Coval's voice AI benchmarks. The live board shows something more useful: three companies serve the same Qwen3-TTS 1.7B weights and land at 2nd, 4th and 28th of 28, a 10.8x spread in time-to-first-audio.
Breeze TTS 2 is the best-sounding open-weights voice model of the four here and the one you are least likely to be allowed to use. In open voice AI, license and hardware decide the pick long before audio quality does.
Microsoft AI released MAI-Transcribe-2, a speech-to-text model at $0.10 per audio hour across 60 languages, with speaker diarization and word-level timestamps built in.
Gemini 3.5 Transcribe, GPT-4o Transcribe, Whisper, and Cohere Transcribe compared on accuracy, languages, streaming, diarization, and price so you can pick the right speech-to-text model.
Descript shipped inline tone tags for ElevenLabs V3 speakers and a connector for Claude and ChatGPT on June 11, 2026, pushing further into directable AI editing.
ElevenLabs Music v2 and Suno both make broadcast-ready tracks. The real 2026 choice is licensing risk versus creative control. Here is how they compare.
Scribix is a browser-based AI transcription tool that turns audio and video into editable text with speaker labels, word-level timestamps, and SRT or VTT export.
NVIDIA released Nemotron 3.5 ASR on June 4: an open 600M streaming speech model covering 40 language-locales with sub-100ms latency for voice agents.
Google Magenta RealTime 2 runs live music generation locally on Mac and Windows, producing instrument tracks in real time from text prompts.
Boson AI has released Higgs Audio v3 TTS 4B, a 4-billion-parameter text-to-speech model built for voice chat. It supports 100 languages with zero-shot voice cloning and inline emotion control tokens.