UK BUSINESS AI UNIVERSE: Labs: Speech and audio AI

AI activity, tracked by AI

4 labs · 40 models and products · Companies

by Alex Connock

Experimental site researched using Gen AI.
Based on public sources. Hallucination & omission likely.

BETAwww.librairie.ai

← Back to the Labs tiles

Speech and audio AI

4 labs · 40 models and products Theme: Companies · Generated from this page's data. Click a name to open its panel on the Labs page. Records resting on one source are tagged "one source".

Speech and audio AI turns text into spoken words, copies a voice from a sample, turns speech into text, and generates music and sound effects. An early example of the modern approach is WaveNet, a 2016 DeepMind paper on generating raw audio waveforms. ElevenLabs, which has teams in London among other cities, builds models that turn text into speech, clone voices and run voice agents, used for audiobooks, dubbing and customer calls (per the site). The tile also lists Google DeepMind's Lyria music models and Stability AI's Stable Audio models.

Context

  1. DeepMind's WaveNet introduced a neural network that generates raw audio waveforms. 12 September 2016 arxiv.org/abs/1609.03499This paper introduces WaveNet, a deep neural network for generating raw audio waveforms.
  2. ElevenLabs raised $500 million in a Series D led by Sequoia, at an $11 billion valuation (reported). 4 February 2026 elevenlabs.io/blog/series-dThis $500 million funding round values ElevenLabs at $11 billion, more than 3x its valuation one year ago.
  3. ElevenLabs' co-founder Piotr Dabkowski says it builds models across text to speech, transcription and music. 4 February 2026 elevenlabs.io/blog/series-dbuilding foundational models across the full audio stack - text to speech, transcription, music
  4. ElevenLabs says it is expanding across London, New York and other cities. 4 February 2026 elevenlabs.io/blog/series-dcontinuing its international expansion across London, New York
  5. Stability AI released Stable Audio Open, an open audio-generation model. 19 July 2024 arxiv.org/abs/2407.14358Our evaluation shows that the model's performance is competitive with the state-of-the-art across various metrics.

Key papers

Labs Round 8 (31 Oct 2026): introduction and facts as the researchers wrote them; "(per the site)" marks a sentence resting on this site's own description of that lab.

RecordWhat it isPlaceLinksConfidence
Cue (forced alignment model)Chips, models and productsA model that lines up each character of a transcript with the exact moment it is spoken in a recording. released (API; inside Scribe)Key facts Published 28 Oct 2026. Runs inside every Scribe transcription and via the Forced Alignment API. Figures (28 Oct 2026): average timing error 179 ms on full recordings (open-source tools 283 ms to 6,000+ ms); 92.7% of sung words within 0.3 s (next best 82.0%); one hour of audio aligned in about half a second on a production GPU.Made by ElevenLabs · 2026-10-28Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/cue-aligner · elevenlabs.io/blog/category/researchlab's own pages
Dubbing v2Chips, models and productsAn AI dubbing model that translates speech into 90+ languages while keeping the speaker's tone and pacing. released (product)Key facts Released 28 May 2026 in ElevenCreative and ElevenProductions; API "coming soon" at launch; 7-day free trial. Figure: 90+ languages. Earlier work: "The Road to Real-time Dubbing" (2 Nov 2023).Made by ElevenLabs · 2026-05-28Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/introducing-dubbing-v2 · elevenlabs.io/blog/category/researchlab's own pages
Eleven English v1 (eleven_monolingual_v1)Chips, models and productsElevenLabs' first text-to-speech model, trained only on English. deprecatedKey facts Described as the company's "very first model"; trained on an English-only dataset; marked deprecated and due for removal; the 24 Nov 2022 blog describes the beta model (unnamed) as trained on "over 500k hours" of dataMade by ElevenLabs · — (first model; release date not stated)Versions Eleven English v1 → Turbo v2 → Turbo v2.5Londonhelp.elevenlabs.io/hc/en-us/articles/17883183930 · elevenlabs.io/blog/thefirstaithatcanlaughone source
Eleven Flash (v2 and v2.5)Chips, models and productsElevenLabs' fastest text-to-speech model, built for voice agents that must answer in real time. released (API)Key facts Released 18 Dec 2024. Variants: eleven_flash_v2 (English only) and eleven_flash_v2_5 (32 languages). Figures: 75ms plus application and network latency (18 Dec 2024); models page (read 30 Oct 2026) gives ~100ms median for v2.5 and ~135ms for v2; v2.5 about 50% lower price per character.Made by ElevenLabs · 2024-12-18Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/meet-flash · elevenlabs.io/docs/overview/modelslab's own pages
Eleven Multilingual v1Chips, models and productsElevenLabs' first multilingual text-to-speech model, adding seven languages to its English voice model. supersededKey facts Announced 27 Apr 2023. Languages added (as published): French, German, Hindi, Italian, Polish, Portuguese, Spanish. Built on in-house research. Still listed in the character-limits table of the models page (read 30 Oct 2026); not in the main model list. Succeeded by Multilingual v2 (22 Aug 2023).Made by ElevenLabs · 2023-04-27Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/eleven-multilingual-v1 · elevenlabs.io/blog/category/researchone source
Eleven Multilingual v2Chips, models and productsA text-to-speech model that speaks 29 languages, released as ElevenLabs came out of beta. released (API; previous generation)Key facts Released 22 Aug 2023. Figures: 29 languages; up to 10,000 characters (~10 min) per request (models page, read 30 Oct 2026). Status on models page: "previous generation". Access: API (model ID eleven_multilingual_v2).Made by ElevenLabs · 2023-08-22Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/eleven-multilingual-v2 · elevenlabs.io/docs/overview/modelslab's own pages
Eleven MusicChips, models and productsA model that writes full songs from plain-language prompts, with or without vocals, cleared for commercial use. released (product and API)Key facts Launched 5 Aug 2025; built with labels, publishers and artists; cleared for most commercial uses (film, TV, podcasts, ads, games). Version history (models page, read 30 Oct 2026): music_v1 → Music v2 (26 May 2026) → Music v2.5 ("Our most advanced music model.").Made by ElevenLabs · 2025-08-05Versions music_v1 → Music v2 (26 May 2026) → Music v2.5 ("Our most advanced music model.")Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/eleven-music-is-here · elevenlabs.io/blog/category/researchlab's own pages
Eleven Turbo v2 (eleven_turbo_v2)Chips, models and productsElevenLabs' first low-latency English speech model, built for real-time use such as phone systems. deprecated (replace with Flash v2)Key facts Announced 9 Nov 2023; about 400 ms latency, over twice as fast as V1 models; output option mulaw 8kHz for VoIP; English only; 0.5 credits per character; now deprecated, replaced by Flash v2 (BIZ-LAB-7024); access: APIMade by ElevenLabs · 2023-11-09Londonelevenlabs.io/blog/turbo-v2-is-here · elevenlabs.io/docs/overview/modelslab's own pages
Eleven Turbo v2.5Chips, models and productsA low-latency text-to-speech model for real-time conversation in 32 languages; now replaced by Flash. deprecatedKey facts Released 19 Jul 2024. Figures (19 Jul 2024): 32 languages; non-English languages 3x faster, English 25% faster. Access: API. Models page (read 30 Oct 2026) lists eleven_turbo_v2_5 and eleven_turbo_v2 as deprecated, replaced by Flash v2.5 and Flash v2.Made by ElevenLabs · 2024-07-19Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/introducing-turbo-v25 · elevenlabs.io/docs/overview/modelslab's own pages
Eleven v3Chips, models and productsAn expressive text-to-speech model with emotion tags and multi-speaker dialogue in 70+ languages. released (generally available; previous generation)Key facts Alpha research preview 3 Jun 2025; generally available 2 Feb 2026. Figures (2 Feb 2026): "68% reduction in errors"; error rate 15.3% → 4.9% across 27 categories in 8 languages; new version preferred 72% of the time over alpha. Models page (read 30 Oct 2026): 70+ languages, 5,000 characters, "previous generation"; Eleven v3 Conversational ~165ms. Succeeded by Eleven v4 (28 Sep 2026).Made by ElevenLabs · 2025-06-03Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/eleven-v3 · elevenlabs.io/blog/eleven-v3-is-now-generally-avlab's own pages
Eleven v4 and Eleven v4 TurboChips, models and productsElevenLabs' newest text-to-speech models, producing emotional speech in 90+ languages; the Turbo version is built for live agents. released (API and products)Key facts Released 28 Sep 2026 in ElevenAgents, ElevenCreative and ElevenAPI. Figures (28 Sep 2026): 90+ languages; Turbo median inference ~100ms and time to first speech ~150ms; preferred by ~75% of listeners in blind head-to-head tests; ranked #1 by Artificial Analysis (as claimed); instant voice clones from 10 seconds of audio. Models page: 10,000 characters per request. Predecessor: Eleven v3.Made by ElevenLabs · 2026-09-28Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/eleven-v4 · elevenlabs.io/docs/overview/modelslab's own pages
ElevenLabsCompaniesBuilds AI models that turn text into speech, clone voices and run voice agents, used for audiobooks, dubbing and customer calls.
Profile
Founded
2022 (Series D post; first funding round announced with Eleven v1 in 2022 per tender post)
Founders
Mati Staniszewski — Co-Founder; Piotr Dabkowski — Co-Founder
Chief executive
null (not stated on pages opened)
Other leaders
Ethan Tandowsky — Chief Financial Officer (joined 8 Sep 2026); Ashley Kramer — Chief Revenue Officer (joined 2 Sep 2026); Andrew Reed (Sequoia) — joined board with Series D (Feb 2026)
Hq
London (England: London) — blog post of 8 Jun 2026 says ElevenLabs "expands London HQ"; other_offices: New York, San Francisco, Warsaw, Dublin, Tokyo, Seoul, Singapore, Bengaluru, Sydney, São Paulo, Berlin, Paris, Mexico City (Feb 2026); later launches in Canada (Jul 2026), Spain (Apr 2026), Australia and New Zealand (May 2026); California expansion "creating 173 high-paying jobs" (Jun 2026)
Headcount
over 800 — 30 Sep 2026 — https://elevenlabs.io/blog/tender-22bn; 530 across more than 50 countries — 5 May 2026 — https://elevenlabs.io/blog/500m-arr-and-new-investors (UK headcount not published)
What it does
AI audio company. It builds models for lifelike speech, speech-to-text, music and sound effects, and sells them through ElevenAgents (voice and chat agents for businesses), ElevenCreative (tools for speech, music, sound, image and video) and ElevenAPI (developer access).
Research focus
"AI audio foundation models"; emotional conversational models, dubbing and "audio general intelligence" (Series D post)
Models and products
Eleven Realtime — voice-agent model — 20 Oct 2026; Eleven v4 — "our most emotive model" — 28 Sep 2026; Eleven v4 Turbo — named Sep 2026; ElevenLabs CLI v1 — 24 Aug 2026; Procedures in ElevenAgents — 30 Jun 2026; Ads Engine in ElevenCreative — 22 Jun 2026; Flows Agent in ElevenCreative — 4 Jun 2026; Dubbing v2 — 28 May 2026; Music v2 — 26 May 2026; ElevenMusic — 29 Apr 2026; Eleven v3 generally available — 2 Feb 2026; Scribe v2 — 9 Jan 2026; Scribe v2 Medical — named Sep 2026; Eleven v1 — 2022
Customers or partners named
Universal Music Group — multi-year strategic agreement — 10 Sep 2026 — https://elevenlabs.io/blog; UK Government — voice AI for public services — 8 Jun 2026 — https://elevenlabs.io/blog; NVIDIA (NVentures), Santander — enterprise investors — 5 May 2026 — https://elevenlabs.io/blog/500m-arr-and-new-investors
Funding total as published
781,000,000 USD "across five rounds" — as of 4 Feb 2026 — https://elevenlabs.io/blog/series-d
Valuation reported
11,000,000,000 USD — 4 Feb 2026 — https://elevenlabs.io/blog/series-d; 22,000,000,000 USD — 30 Sep 2026 (employee tender) — https://elevenlabs.io/blog/tender-22bn (company only)
Rounds found
2026-02-04 — Series D — 500,000,000 USD — Sequoia Capital — Andreessen Horowitz, ICONIQ, Lightspeed, Evantic, BOND, BroadLight, NFDG, Valor Capital, AMP Coalition, Smash Capital; 2026-05-05 — Series D third close (amount not stated) — no lead named — BlackRock, Wellington, D.E. Shaw, Schroders, NVentures, Santander, T.Capital, 30+ creative-industry individuals, Robinhood Ventures retail access; separate $100m tender closed; 2026-09-30 — employee tender — 300,000,000 USD — Wellington, T. Rowe Price — BDT & MSD, EQT, GIC, Goldman Sachs, OTPP, Sapphire Ventures (new); Alkeon, a16z, D.E. Shaw, DISRUPTIVE, Evantic, ICONIQ, Lightspeed (existing); 2026-06-18 — "Poland invests in ElevenLabs" (post title only; amount not opened)
Uk government links
Partnership with the UK Government to bring voice AI to public services — 8 Jun 2026; AISI voice-AI security work already on site (BIZ-SAFE-2087)
Status now
Valued at $22bn after a $300m employee tender closed in September 2026; over 800 staff; ARR passed $500m in the first four months of 2026.

Labs Round 7 deep profile (30 Oct 2026), as the researchers wrote it; confidence: single_source (company pages only).

Sources

  1. elevenlabs.io/blog/series-d
  2. elevenlabs.io/blog/series-d
  3. elevenlabs.io/blog/series-d
  4. elevenlabs.io/blog/tender-22bn
  5. elevenlabs.io/blog/tender-22bn
  6. elevenlabs.io/blog/500m-arr-and-new-investors
  7. elevenlabs.io/about
  8. elevenlabs.io/blog
Models and products (21)

Version chainsmusic_v1 → Music v2 (26 May 2026) → Music v2.5 ("Our most advanced music model.")Eleven English v1 → Turbo v2 → Turbo v2.5

  1. Cue (forced alignment model) 2026-10-28 released (API; inside Scribe)A model that lines up each character of a transcript with the exact moment it is spoken in a recording.
  2. Eleven v4 and Eleven v4 Turbo 2026-09-28 released (API and products)ElevenLabs' newest text-to-speech models, producing emotional speech in 90+ languages; the Turbo version is built for live agents.
  3. Dubbing v2 2026-05-28 released (product)An AI dubbing model that translates speech into 90+ languages while keeping the speaker's tone and pacing.
  4. Music v2 (and Music v2.5) 2026-05-26 released (product)ElevenLabs' second music model, with better vocals, instruments and song structure and multi-language lyrics.
  5. Scribe v2 (and Scribe v2 Medical) 2026-01-09 released (API)ElevenLabs' second-generation transcription model for bulk transcripts, subtitles and captions; a medical variant handles clinical audio.
  6. Scribe v2 Realtime 2025-11-11 released (API)A live speech-to-text model for voice agents, meetings and captions.
  7. Eleven Music 2025-08-05 released (product and API)A model that writes full songs from plain-language prompts, with or without vocals, cleared for commercial use.
  8. Voice Design v3 2025-06-25 released (product)A tool that creates a new synthetic voice from a written description and offers three options.
  9. Eleven v3 2025-06-03 released (generally available; previous generation)An expressive text-to-speech model with emotion tags and multi-speaker dialogue in 70+ languages.
  10. Scribe (v1) 2025-02-26 deprecatedElevenLabs' first speech-to-text model, transcribing 99 languages with timings and speaker labels.
  11. Eleven Flash (v2 and v2.5) 2024-12-18 released (API)ElevenLabs' fastest text-to-speech model, built for voice agents that must answer in real time.
  12. Eleven Turbo v2.5 2024-07-19 deprecatedA low-latency text-to-speech model for real-time conversation in 32 languages; now replaced by Flash.
  13. Eleven Turbo v2 (eleven_turbo_v2) 2023-11-09 deprecated (replace with Flash v2)ElevenLabs' first low-latency English speech model, built for real-time use such as phone systems.
  14. Eleven Multilingual v2 2023-08-22 released (API; previous generation)A text-to-speech model that speaks 29 languages, released as ElevenLabs came out of beta.
  15. Eleven Multilingual v1 2023-04-27 supersededElevenLabs' first multilingual text-to-speech model, adding seven languages to its English voice model.
  16. Voice Design (first generative voice model, in Voice Lab) 2023-01-11 superseded (product)ElevenLabs model that creates new synthetic voices from scratch rather than copying a real speaker.
  17. Eleven Realtime v1 (Mini and Max) no date released (API; date not published)Complete speech-to-speech voice-agent models: Mini tuned for speed and natural turn-taking, Max for complex multi-step tasks with tool calls.
  18. SFX v3 (sound effects model) no date limited access (date not published)A model that makes sound effects from text prompts, with better prompt-following and looping.
  19. Eleven English v1 (eleven_monolingual_v1) — (first model; release date not stated) deprecatedElevenLabs' first text-to-speech model, trained only on English.
  20. Speech to Speech voice changer models (English STS v1 and v2; Multilingual STS v2) — (release dates not stated) released (API)ElevenLabs models that turn one person's recorded speech into another voice while keeping the delivery.
  21. Text to Sound Effects v2 (eleven_text_to_sound_v2) — (release date not stated) released (API; default model)ElevenLabs model that makes sound effects from a written description.
Key facts and news (8)
Status now
Valuation reported at $22bn after a $300m employee tender offer led by Wellington and T. Rowe Price (30 Sep 2026); over 800 staff; London HQ to triple in size and UK headcount to double to 200 (8 Jun 2026) one source elevenlabs.io/blog/tender-22bn
Valuation (reported)
$22bn (30 Sep 2026, employee tender offer; described as double the February 2026 Series D valuation) one source elevenlabs.io/blog/tender-22bn
Users and customers
Enterprise is 55% of revenue; agents handle over 15 million conversations a week; used by five of the top ten tech companies, five of the top ten insurers and four of the top ten telecoms (30 Sep 2026); Revolut uses its voice agents for 4 million customers in 30 languages (8 Jun 2026); crossed $500m ARR (5 May 2026) one source elevenlabs.io/blog/tender-22bn
Headcount
Over 800 (30 Sep 2026); UK headcount planned to double to 200 across research, engineering and sales (8 Jun 2026) one source elevenlabs.io/blog/uk-mou-and-expansion
Models released
Deprecated per model docs: eleven_monolingual_v1 (English v1), eleven_multilingual_v1, eleven_turbo_v2, eleven_turbo_v2_5, scribe_v1; current docs list eleven_v4 and eleven_v4_turbo (90+ languages) as newest TTS one source elevenlabs.io/docs/overview/models

News

Labs Round 8 (31 Oct 2026), from the researchers' rows; one-source items are marked.

Soho, London (W1F 0DQ)Companies House · elevenlabs.io/de/blog/series-d · tech.eu/2026/02/04/elevenlabs-raises-500m-says-b · uktech.news/ai/elevenlabs-valued-at-11bn-after-s · elevenlabs.io/blog/tender-22bnverified
Gemini 3.1 Flash Live and Gemini 3.1 Flash TTSChips, models and productsGemini voice models: Flash Live for real-time spoken conversation and Flash TTS for directed, expressive speech. releasedKey facts Theme: Speech and audio AI. Flash Live released 26 Mar 2026; Flash TTS released 15 Apr 2026 with audio tags; model card "Gemini 3.1 Flash Audio (Flash Live, TTS)" updated 15 Apr 2026. Earlier: Gemini 2.5 native audio (Jun 2025); improved audio models (12 Dec 2025). Later: Gemini 3.8 audio (Sep 2026).Made by Google DeepMind · 2026-03-26London (EC2M 4AG) (via Google DeepMind)deepmind.google/blog/rss.xml · deepmind.google/blog/rss.xmllab's own pages
Gemini 3.5 Live Translate and Gemini 3.5 TranscribeChips, models and productsGemini speech models for near real-time spoken translation and for speech-to-text transcription. releasedKey facts Theme: Speech and audio AI. Live Translate 9 Jun 2026 (AI Studio, Google Translate, Google Meet); Transcribe 26 Aug 2026; model card "Gemini 3.5 Audio (Live Translate, Transcribe, Transcribe Live)" updated 26 Aug 2026.Made by Google DeepMind · 2026-06-09London (EC2M 4AG) (via Google DeepMind)deepmind.google/blog/rss.xml · deepmind.google/blog/rss.xmllab's own pages
Gemini 3.8 audio models (Live, Live Extended Thinking, TTS, Live Avatar)Chips, models and productsGemini's September 2026 real-time voice, text-to-speech and talking-avatar models. releasedKey facts Theme: Speech and audio AI. 3.8 Live and Live Extended Thinking 15 Sep 2026; 3.8 text-to-speech 23 Sep 2026; 3.8 Live with Live Avatar 24 Sep 2026; model card "Gemini 3.8 Audio (Live, Live Extended Thinking, Flash TTS, Flash-Lite TTS)" updated 24 Sep 2026.Made by Google DeepMind · 2026-09-15London (EC2M 4AG) (via Google DeepMind)deepmind.google/blog/rss.xml · deepmind.google/blog/rss.xmllab's own pages
Lyria (first version) and Dream TrackChips, models and productsGoogle DeepMind's first Lyria music-generation model, launched with a YouTube Shorts experiment that made songs in artists' voices. released (limited: YouTube experiment)Key facts Announced 16 Nov 2023; Dream Track experiment with nine artists for a limited set of creators; all Lyria output watermarked with SynthID; access: limited product testMade by Google DeepMind · 2023-11-16Versions Lyria → Lyria 2Londondeepmind.google/blog/transforming-the-future-of- · deepmind.google/blog/2023-a-year-of-groundbreakilab's own pages
Lyria 2 (and Music AI Sandbox update)Chips, models and productsSecond Lyria music model, offered to musicians and producers in the US through Google's Music AI Sandbox. released (waitlist, US)Key facts Announced 24 Apr 2025 within Music AI Sandbox; access widened to more US musicians, producers and songwriters via waitlist; output watermarked with SynthID; Lyria RealTime named in same post (already listed as BIZ-LAB-7172)Made by Google DeepMind · 2025-04-24Versions Lyria → Lyria 2Londondeepmind.google/blog/music-ai-sandbox-now-with-n · deepmind.google/models/lyria/one source
Lyria 3Chips, models and productsGoogle DeepMind's music generation model that turns a text prompt into a song; it made music creation available in the Gemini app. released (Gemini app, Vertex AI, AI Studio, YouTube, API)Key facts Theme: Speech and audio AI. Released 18 Feb 2026 ("Gemini can now create music"); card updated March 2026; outputs watermarked with SynthID; compared against Lyria 2 (no numeric results). Channels: Gemini app, Google Cloud/Vertex AI, AI Studio, Google Vids, YouTube, Producer AI, Music AI Sandbox.Made by Google DeepMind · 2026-02-18London (EC2M 4AG) (via Google DeepMind)deepmind.google/models/model-cards/lyria-3/ · deepmind.google/blog/rss.xmllab's own pages
Lyria 3 ProChips, models and productsA Lyria music model that writes longer tracks with song structure. releasedKey facts Theme: Speech and audio AI. Released 25 Mar 2026; Lyria brought to more Google products.Made by Google DeepMind · 2026-03-25London (EC2M 4AG) (via Google DeepMind)deepmind.google/blog/rss.xmlone source
Lyria 3.5Chips, models and productsGoogle DeepMind's newest music model, with better musicality, lyrics and vocals. released (Google Flow Music; Gemini)Key facts Theme: Speech and audio AI. Launched 29 Jul 2026 in Google Flow Music; card same date. Lyria page (read 30 Oct 2026): tracks "up to 3 minutes long"; also lists Lyria RealTime (interactive real-time music) and Magenta RealTime (open Lyria model).Made by Google DeepMind · 2026-07-29London (EC2M 4AG) (via Google DeepMind)deepmind.google/models/lyria/ · deepmind.google/models/model-cards/lab's own pages
Magenta RealTime and Lyria RealTimeChips, models and productsReal-time interactive music generation models; Magenta RealTime is the open Lyria model. released (Magenta RealTime open; dates not on page)Key facts Theme: Speech and audio AI. Listed on the Lyria page (read 30 Oct 2026) with no dates. Earlier: Music AI Sandbox with new features and broader access (April 2025 index).Made by Google DeepMindLondon (EC2M 4AG) (via Google DeepMind)deepmind.google/models/lyria/ · deepmind.google/blog/page/7/one source
Melia (Speechmatics multilingual speech-to-text model)Chips, models and productsSpeechmatics' speech-to-text model that follows speakers who switch between languages within one recording. live (production preview, batch transcription)Key facts 55+ languages in one model (17 Jun 2026); API name melia-1; pricing from $0.129 per hour with 10 hours a month free (17 Jun 2026); wins on FLEURS against Deepgram in 91% of languages, Microsoft 91%, AssemblyAI 77% (vendor figures, 17 Jun 2026)Made by Speechmatics · 2026-06-17Cambridge (CB4 0AE) (via Speechmatics)speechmatics.com/company/articles-and-news/intro · speechmatics.com/lab's own pages
Multi-speaker dialogue generation (behind NotebookLM Audio Overviews)Chips, models and productsGoogle DeepMind speech technology that writes out two-person podcast-style conversations as audio, used in NotebookLM Audio Overviews. live (inside Google products)Key facts Post 30 Oct 2024; produces 2 minutes of dialogue in under 3 seconds on a single TPU v5e chip, over 40 times faster than real time; underpins NotebookLM Audio Overviews and Illuminate; model not named; access: inside Google productsMade by Google DeepMind · 2024-10-30Londondeepmind.google/blog/pushing-the-frontiers-of-auone source
Music v2 (and Music v2.5)Chips, models and productsElevenLabs' second music model, with better vocals, instruments and song structure and multi-language lyrics. released (product)Key facts Released 26 May 2026 on ElevenMusic and ElevenCreative; API "coming soon" at launch. Pricing for Music v1 and v2 cut by up to 50% (API) and up to 40% (self-serve). Cleared for commercial use. Models page (read 30 Oct 2026) adds Music v2.5 (music_v2_5), undated.Made by ElevenLabs · 2026-05-26Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/introducing-music-v2 · elevenlabs.io/docs/overview/modelslab's own pages
Oak-1 (Speechmatics medical speech-to-text model)Chips, models and productsSpeechmatics' medical transcription model for clinical consultations where several people speak and the language may change. liveKey facts Average 91% medical-term recall across 11 benchmarked languages vs 74% for eight other providers (vendor figures, 14 Oct 2026); Arabic–English code-switching mixed error rate 15.1% vs 33.2% next best; transcribes an hour of multilingual audio in about 18 seconds in batchMade by Speechmatics · 2026-10-14Cambridge (CB4 0AE) (via Speechmatics)speechmatics.com/company/articles-and-news/meet- · speechmatics.com/lab's own pages
Owl (PolyAI speech recognition model)Chips, models and productsPolyAI's own speech-to-text model, built to hear callers accurately over phone lines with noise and accents. live (inside PolyAI platform)Key facts Announced 3 Jun 2025; word error rate 0.122, stated as most accurate of the models PolyAI tested; PolyAI says about 70% of its voice agents' errors come from speech recognition; used in PolyAI Agent Studio; access: within PolyAI platform (no standalone release stated)Made by PolyAI · 2025-06-03Londonpoly.ai/blog/owl-speech-recognition-model-polyai · poly.ai/technologylab's own pages
PapercuphistoricalCompanies · also on Start-upsMade software that dubbed video into other languages with synthetic voices checked by human editors; sold its technology to RWS in 2025 and is in liquidation. wound_downWhy it is here: historical (acquisition and wind-down 2025 already on site; lead reports it in liquidation) — reason: Papercup researchers built their own text-to-speech prosody models and datasets
Selected papers (2)
  1. Controllable Prosody Generation with Partial Inputs (Iliescu, Ram Mohan, Teh, Hodari; Papercup Technologies) arXiv preprint · Mar 2023
  2. ADEPT: A Dataset for Evaluating Prosody Transfer (Torresquintero et al.; Papercup Technologies Ltd and University of Edinburgh) arXiv preprint · Jun 2021
LondonCompanies House · rws.com/about/news/2025/rws-acquires-papercups-i · find-and-update.company-information.service.gov. · uktech.news/ai/papercup-funding-16m-20220610 · tvbeurope.com/artificial-intelligence/papercup-iverified
Scribe (v1)Chips, models and productsElevenLabs' first speech-to-text model, transcribing 99 languages with timings and speaker labels. deprecatedKey facts Released 26 Feb 2025. Figures (26 Feb 2025): 99 languages; FLEURS and Common Voice tests; Italian 98.7% and English 96.7% accuracy as published. Access: Speech to Text API and dashboard. Models page (read 30 Oct 2026) lists scribe_v1 as deprecated, replaced by scribe_v2.Made by ElevenLabs · 2025-02-26Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/meet-scribe · elevenlabs.io/docs/overview/modelslab's own pages
Scribe v2 (and Scribe v2 Medical)Chips, models and productsElevenLabs' second-generation transcription model for bulk transcripts, subtitles and captions; a medical variant handles clinical audio. released (API)Key facts Released 9 Jan 2026. Claims "lowest word error rate recorded on industry-standard benchmarks" (FLEURS named). Up to 100 key terms; up to 56 entity categories. Models page (read 30 Oct 2026): 90+ languages; diarisation up to 32 speakers; Scribe v2 Medical is batch-only and HIPAA-eligible. Predecessor: Scribe v1.Made by ElevenLabs · 2026-01-09Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/introducing-scribe-v2 · elevenlabs.io/docs/overview/modelslab's own pages
Scribe v2 RealtimeChips, models and productsA live speech-to-text model for voice agents, meetings and captions. released (API)Key facts Released 11 Nov 2025. Figures (11 Nov 2025): under 150 ms latency; 90 languages; 93.5% accuracy across 30 common European and Asian languages. Access: API and ElevenLabs Agents.Made by ElevenLabs · 2025-11-11Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/introducing-scribe-v2-realtim · elevenlabs.io/docs/overview/modelslab's own pages
SFX v3 (sound effects model)Chips, models and productsA model that makes sound effects from text prompts, with better prompt-following and looping. limited access (date not published)Key facts Models page (read 30 Oct 2026): eleven_text_to_sound_v3 in limited access, not in every region; eleven_text_to_sound_v2 remains the Sound Effects API default. No release date published there.Made by ElevenLabsSoho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/docs/overview/modelsone source
Speech to Speech voice changer models (English STS v1 and v2; Multilingual STS v2)Chips, models and productsElevenLabs models that turn one person's recorded speech into another voice while keeping the delivery. released (API)Key facts Model ids eleven_english_sts_v1, eleven_english_sts_v2 (English only) and eleven_multilingual_sts_v2 (29 languages); 10,000-character limit listed for English STS; ElevenLabs blogged on voice conversion research on 6 Sep 2022; access: APIMade by ElevenLabs · — (release dates not stated)Londonelevenlabs.io/docs/overview/models · elevenlabs.io/blog/voice-conversionone source
SpeechmaticsCompanies · also on Software and toolsMakes speech-to-text software that understands many languages, sold to other companies to build into their products.speech recognition
Profile
Founded
2009 (UKTN 2021 Cambridge list)
Founders
Tony Robinson — founder (UKTN 2021)
Chief executive
Katy Wigdahl — from null (CEO in June 2022 per UKTN; still CEO April 2026 per company); John Milliken was chief executive in October 2019
Other leaders
Lauren King — Chief Marketing Officer (2026-01); Jonathan Klahr (Susquehanna Growth Equity) — board member (from 2022-06)
Hq
Cambridge (England: East of England); company boilerplate says "headquartered in Cambridge and London"
Other offices
London; Boston (Outside the UK); Chennai, India (Outside the UK); Czech Republic (Outside the UK) — offices as listed by UKTN 2022-06
What it does
Speech-to-text software sold through an API. It turns live or recorded speech in 55+ languages into text, in the cloud, on a company's own servers or on the device itself, and is used for voice agents, captioning and medical note-taking.
Products
Medical Model for Real-time (medical speech-to-text; 93% accuracy, 96% medical keyword recall) — preview 2025-09-15; On Device speech-to-text model — 2025 (in Adobe Premiere 2026-04-21); ForceEndOfUtterance (final transcript within 250ms) — by 2026-02; English-Arabic bilingual model — announced for early 2026
Customers named
Deloitte UK, 3Play Media, Veritone — https://uktech.news/ai/speechmatics-series-b-20220628; AI-Media, Content Guru, Nordhealth Therapy — https://www.speechmatics.com/company/articles-and-news/speechmatics-sets-record-in-medical-speech-to-text-with-93-percent-accuracy; Adobe (Premiere) — https://www.speechmatics.com/company/articles-and-news/adobe-and-speechmatics-deliver-cloud-grade-speech-recognition-on-device-for-premiere
Partnerships
Boost.ai — enterprise voice AI for regulated industries in Europe, extending to the US — 2026-02-12; Sully.ai — autonomous healthcare agents and scribes on NVIDIA infrastructure — 2026-01-12; Adobe — on-device speech-to-text in Premiere — 2026-04-21; NVIDIA — medical model optimised through NVIDIA Triton Inference Server — 2025-09-15; LiveKit and Pipecat — voice-agent platform partners — by 2025-09; Edvak EHR — speech layer for clinical documentation in an AI-native health record system — 2026-02-17; Ambarella — speech recognition on Ambarella CVflow AI chips for offline edge devices, shown at CES — 2025-01-02; Barometer — transcription and brand-safety data for podcast advertising — 2025-04-22
Usage or revenue as published
real-time usage up 4x 2024 to 2025; voice-agent usage up 9x in 2025; medical model usage up 15x year on year — 2026-01-07 — https://www.speechmatics.com/company/articles-and-news/speechmatics-in-2025-the-numbers-that-shaped-voice-ais-breakthrough-year; medical models available in English, German, French, Swedish, Norwegian, Danish, Finnish and Arabic — 2026-02-17 — https://www.speechmatics.com/company/articles-and-news/speechmatics-and-edvak-ehr-partner-to-make-voice-safe-for-clinical-automation
Funding total as published
10.2m GBP — as of 2021-05 (before the 2022 Series B) — https://www.uktech.news/news/promising-deep-tech-startups-of-cambridge-2021-20210510
Rounds found
2019-10-21 — Series A — 6.35m GBP — AlbionVC — IQ Capital, angel investors; 2022-06-28 — Series B — 62m USD (50.3m GBP) — Susquehanna Growth Equity — AlbionVC, IQ Capital (already DEAL-SW-053)
Status now
Active; Cambridge-based speech-to-text company led by Katy Wigdahl, announcing healthcare and voice-agent partnerships through 2026.
Confidence
verified (CEO on trade and primary pages); funding rounds single_source

Round 6 deep profile (30 Oct 2026), as the researchers wrote it; sources are listed in the record panel.

Models and products (3)
  1. Oak-1 (Speechmatics medical speech-to-text model) 2026-10-14 liveSpeechmatics' medical transcription model for clinical consultations where several people speak and the language may change.
  2. Linden-1 (Speechmatics Agent STT model) 2026-09-17 liveSpeechmatics' fast speech-recognition model built for AI voice agents, returning who said what in real time.
  3. Melia (Speechmatics multilingual speech-to-text model) 2026-06-17 live (production preview, batch transcription)Speechmatics' speech-to-text model that follows speakers who switch between languages within one recording.
Why it is here: reason: builds its own speech-recognition models (Melia, Linden-1, Oak-1)
Key facts and news (4)
Models released
Melia (melia-1, 17 Jun 2026); Linden-1 (Agent STT, 17 Sep 2026); Oak-1 (medical, 14 Oct 2026) speechmatics.com/
Founders
Dr Tony Robinson — founder (neural-network speech recognition pioneer at Cambridge University in the 1980s) speechmatics.com/company/about-speechmatics
Chief executive
Katy Wigdahl (CEO); board includes Jonathan Klar (Susquehanna Growth Equity), Robert Whitby-Smith (Albion VC), Ed Stacey (IQ Capital) one source speechmatics.com/company/about-speechmatics
Awards
Queen's Award for Enterprise (2019); ISO/IEC 27001:2022 accreditation one source speechmatics.com/

Labs Round 8 (31 Oct 2026), from the researchers' rows; one-source items are marked.

Cambridge (CB4 0AE)Companies House · find-and-update.company-information.service.gov. · uktech.news/ai/speechmatics-series-b-20220628 · uktech.news/featured/speechmatics-raises-6-35m-s · uktech.news/ai/speechmatics-series-b-20220628one source
Stability AICompaniesBuilds open-weight generative models that make images, video and audio from text, including Stable Diffusion and Stable Audio, for media, games and music companies.
Profile
Founders
null (not named on pages opened)
Chief executive
Prem Akkaraju — CEO from 25 Jun 2024 (former CEO of Weta Digital)
Chief scientist or research lead
Christian Laforte — CTO (June 2024; previously interim co-CEO from March 2024)
Other leaders
Sean Parker — Executive Chairman (from 25 Jun 2024); Shan Shan Wong — COO (June 2024); Robert Legato — Chief Pipeline Architect (19 Mar 2025); Thomas Laffont (Coatue co-founder) — joined board 25 Aug 2026; Dana Settle (Greycroft) — board; James Cameron — board
Hq
null (2024 and 2026 releases datelined Los Angeles; no HQ stated). Site already holds a 2026 UK office closure (DEAL-LAB-014 / DEAL-LAB-2026).
What it does
Generative AI company that makes open-weight models for images (Stable Diffusion), audio and music (Stable Audio), video, 3D and multi-view camera work, plus enterprise creative tools such as Brand Studio. It now works closely with record labels, a games publisher and an advertising group.
Research focus
open-weight image, video, 3D and audio generation; enterprise creative production
Models and products
Stable Audio 3.0 — open-weight music model family — 20 May 2026; Brand Studio — creative production platform — 8 Apr 2026; Stable Audio 2.5 — enterprise sound production — 10 Sep 2025; Stable Diffusion 3.5 NIM (with NVIDIA) — 12 Aug 2025; Stability AI Solutions — 5 Aug 2025; Stable Video 4D 2.0 — 20 May 2025; Stable Audio Open Small (with Arm) — 14 May 2025; Stable Virtual Camera — multi-view video — 18 Mar 2025
Customers or partners named
Warner Music Group — AI music creation tools — 19 Nov 2025 — https://stability.ai/news-updates; Universal Music Group — strategic alliance on professional AI music tools — 30 Oct 2025 — same url; Electronic Arts — game development partnership — 23 Oct 2025 — same url; Amazon Bedrock — image services — 18 Sep 2025 — same url; NVIDIA — SD 3.5 NIM — 12 Aug 2025 — same url; Arm — 14 May 2025 — same url; AMD — 16 Apr 2025 — same url; joined the Tech Coalition — 11 Feb 2026 — same url
Funding total as published
232,000,000 USD, "inclusive of two equity rounds and convertible notes" — as of 25 Aug 2026 — https://stability.ai/news-updates/stability-ai-latest-funding-backed-by-entertainment-industry-biggest-names
Rounds found
2024-06-25 — initial round under new management — undisclosed — no lead named — Greycroft, Coatue, Sound Ventures, Lightspeed, O'Shaughnessy Ventures, Sean Parker, Prem Akkaraju, Eric Schmidt, Robert Nelsen; 2026-08-25 — Series B — 76,000,000 USD new capital — no lead named — Electronic Arts, Sony Music Group, Universal Music Group, Warner Music Group, AMD Ventures, Pacific Alliance Ventures (new); Coatue, Greycroft, Kadmos Capital, Lightspeed, Mantis Capital, Sound Ventures, WPP (returning); individuals incl. Sean Parker, Eric Schmidt, James Cameron
Status now
Raised a $76m Series B in August 2026 backed by music labels, EA and Sony Music; releases are datelined Los Angeles.

Labs Round 7 deep profile (30 Oct 2026), as the researchers wrote it; confidence: single_source.

Sources

  1. stability.ai/news-updates/stability-ai-latest-fu
  2. stability.ai/news-updates/stability-ai-latest-fu
  3. stability.ai/news-updates/stability-ai-secures-s
  4. stability.ai/news-updates
Models and products (27)

Version chainsStable Audio 2.0 (3 Apr 2024) → Stable Audio Open (5 Jun 2024) → Stable Audio Open Small (14 May 2025) → Stable Audio 2.5 (10 Sep 2025) → Stable Audio 3.0 (20 May 2026)StableLM → Stable LM 2Stable Diffusion v1 → 2.0 → SDXL → SDXL Turbo → SD3 → SD3.5

  1. Stable Audio 3.0 2026-05-20 released (open weights for Small SFX, Small, Medium; Large via API)Stability AI's audio model family for music and sound effects, trained on licensed data, mostly with open weights.
  2. Brand Studio 2026-04-08 released (product)Stability AI's creative production platform that generates content in a company's own brand style.
  3. Stable Audio 2.5 2025-09-10 released (API and enterprise licence)An enterprise audio model that generates brand music and sound in seconds.
  4. Stability AI Solutions 2025-08-05 released (enterprise service)Stability AI's packaged generative AI services for enterprise creative production.
  5. Stable Audio Open Small 2025-05-14 released (open weights)A small open audio model, released with Arm, that generates sound on phones and other devices.
  6. Stable Virtual Camera 2025-03-18 released (model)A model that makes multi-view video from images with 3D camera control.
  7. Stable Point Aware 3D (SPAR3D) 2025-01-08 released (model)A model that generates a complete 3D object from an image and allows real-time editing.
  8. Stable Diffusion 3.5 2024-10-22 released (model)Stability AI's text-to-image model family released in October 2024 and later offered through cloud marketplaces.
  9. Stable Fast 3D 2024-08-01 released (model)A model that makes a 3D asset from a single image quickly.
  10. Stable Video 4D (and SV4D 2.0) 2024-07-24 released (model)A model that turns one video into multi-angle video of a moving object.
  11. Stable Assistant 2024-07-09 released (product)Stability AI's assistant product, through which its image models were offered to users.
  12. Stable Audio Open 2024-06-05 released (open source)An open-source model for making short audio samples and sound effects.
  13. Stable Audio 2.0 2024-04-03 released (product)Stability AI's second-generation music and sound generation model.
  14. Stable Video 3D 2024-03-18 released (model)A model that generates new viewing angles and 3D objects from one image.
  15. TripoSR 2024-03-05 released (model)A model that makes a 3D object from a single image quickly.
  16. Stable Diffusion 3 (incl. SD3 Medium) 2024-02-22 released (API; open weights for Medium)Stability AI's third-generation text-to-image model; the 2-billion-parameter Medium version was released with open weights.
  17. Stable Cascade 2024-02-12 released (model)A Stability AI model announced in February 2024; launch post not opened, so details are not recorded.
  18. Stable LM 2 (1.6B and 12B) 2024-01-19 released (model)Stability AI's second family of small open language models, in 1.6-billion and 12-billion-parameter sizes.
  19. Stable Code 3B 2024-01-16 released (model)A small (3-billion-parameter, per its name) coding language model from Stability AI; launch post titled "Coding on the Edge".
  20. Stable Zero123 (and Stable Zero123C) 13 Dec 2023 (post updated 8 Jan 2024) released (non-commercial; Zero123C commercial for members)Model that generates new views of an object from a single picture, as a step to making 3D objects.
  21. SDXL Turbo (Adversarial Diffusion Distillation) 2023-11-28 released (non-commercial open weights)Fast version of SDXL that makes an image in a single step, close to real time.
  22. Stable Video Diffusion 2023-11-21 research preview (open code and weights)Stability AI's first video model, which turns a single image into a short video clip.
  23. SDXL (Stable Diffusion XL) 2023-07-04 (paper) superseded (open weights)Larger Stable Diffusion model for higher-resolution images, with a second text encoder and an optional refining model.
  24. DeepFloyd IF 28 Apr 2023 (year from page image timestamps) released (research licence)Text-to-image model from Stability AI's DeepFloyd research lab, built in three stages and able to write legible text inside images.
  25. Stable Diffusion (first public release, v1) no date superseded (open weights)The original open text-to-image model that made pictures from typed prompts on an ordinary consumer graphics card.
  26. Stable Diffusion 2.0 no date superseded (open weights)Second major Stable Diffusion release, with a new text encoder, higher default resolution and a 4x upscaler.
  27. StableLM (Alpha, 3B and 7B) no date superseded (open weights)Stability AI's first suite of open language models, released in two small sizes.
Why it is here: reason: trains its own audio models (Stable Audio 3.0 family; DAW plugin powered by its own models, Aug 2026)
Selected papers (12)
  1. Arbor: Explicit Geometric Conditioning for Controllable 3D Asset Generation Stability AI research · 7 October 2026
  2. 4Director: Controlling Video World Models with Rigid 3D Geometry 7 October 2026
  3. SemanTok: Predictable Semantic Tokens for Efficient Autoregressive Video Generation 7 October 2026
  4. Stable-Layers 3 June 2026
  5. Stable Audio 3 and SAME 20 May 2026
  6. Stable Audio 3 arXiv preprint · 18 May 2026The Stable Audio 3 model family generates music and sound in under two seconds on one H200 GPU and can edit or extend recordings.
  7. SD3.5-Flash: Distribution-Guided Distillation of Generative Flows 26 September 2025
  8. Stable Audio Open arXiv preprint · 19 Jul 2024An open-weight text-to-audio model that performed on a par with the best systems on several measures.
  9. Scaling Rectified Flow Transformers for High-Resolution Image Synthesis (Stable Diffusion 3 paper) arXiv preprint · 5 Mar 2024The architecture behind Stable Diffusion 3; the biggest models outperformed the best image generators of the time.
  10. Adversarial Diffusion Distillation (SDXL Turbo method) arXiv preprint · 28 Nov 2023A training method that lets a large image model produce a picture in a single step, in real time.
  11. Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets arXiv preprint · 25 Nov 2023Training on large video datasets produced a text-to-video model competitive with closed-source systems.
  12. SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis arXiv preprint · 4 Jul 2023The SDXL image model clearly outperformed earlier versions of Stable Diffusion; code and weights released on Stability AI's GitHub.
Key facts and news (7)
Status now
Led by CEO Prem Akkaraju (appointed June 2024); $76m new capital in Series B (25 Aug 2026) brought total funding to $232m; Coatue co-founder Thomas Laffont joined the board; announcement datelined Los Angeles one source stability.ai/news-updates/stability-ai-latest-fu
Funding as published (reported)
$232m including two equity rounds and convertible notes (25 Aug 2026) one source stability.ai/news-updates/stability-ai-latest-fu
Users and customers
Partners and backers named: Electronic Arts, Sony Music Group, Universal Music Group, Warner Music Group, WPP, AWS (Amazon Bedrock), NVIDIA, AMD, Arm stability.ai/news
Research focus
3D asset generation, controllable video world models, efficient autoregressive video, image layer decomposition, KV-cache compression, music autoencoders and text-to-audio (2025–26 research posts) stability.ai/research
Models released
Core Models list (updated 20 May 2026) also names SD Turbo, Japanese SDXL, Japanese Stable CLIP, Stable LM Zephyr 3B, Japanese Stable LM 2 1.6B, Japanese Stable VLM and Stable Code Instruct 3B (no dates on page; not added as products — UK build not shown for the Japanese models) one source stability.ai/core-models

News

Labs Round 8 (31 Oct 2026), from the researchers' rows; one-source items are marked.

Los AngelesCompanies House · stability.ai/news-updates/stability-ai-latest-fu · tech.eu/2026/08/25/stabiityai-raises-76m-with-wa · stability.ai/news/stability-ai-secures-significa · find-and-update.company-information.service.gov.verified
Stable Audio 2.0Chips, models and productsStability AI's second-generation music and sound generation model. released (product)Key facts Theme: Speech and audio AI. Announced 3 Apr 2024. Version history (lab posts): Stable Audio 2.0 (3 Apr 2024) → Stable Audio Open (5 Jun 2024) → Stable Audio Open Small (14 May 2025) → Stable Audio 2.5 (10 Sep 2025) → Stable Audio 3.0 (20 May 2026).Made by Stability AI · 2024-04-03Versions Stable Audio 2.0 (3 Apr 2024) → Stable Audio Open (5 Jun 2024) → Stable Audio Open Small (14 May 2025) → Stable Audio 2.5 (10 Sep 2025) → Stable Audio 3.0 (20 May 2026)Los Angeles (via Stability AI)stability.ai/news-updates/category/Productone source
Stable Audio 2.5Chips, models and productsAn enterprise audio model that generates brand music and sound in seconds. released (API and enterprise licence)Key facts Theme: Speech and audio AI. Announced 10 Sep 2025. Figures (10 Sep 2025): tracks up to three minutes; generation in under two seconds on a GPU. Access: StableAudio.com, Stability AI API, partners fal, Replicate, ComfyUI; on-premises under enterprise licence.Made by Stability AI · 2025-09-10Los Angeles (via Stability AI)stability.ai/news-updates/stability-ai-introduce · stability.ai/news-updateslab's own pages
Stable Audio 3.0Chips, models and productsStability AI's audio model family for music and sound effects, trained on licensed data, mostly with open weights. released (open weights for Small SFX, Small, Medium; Large via API)Key facts Theme: Speech and audio AI. Announced 20 May 2026. Variants: Small SFX, Small, Medium (open weights on Hugging Face, Community Licence; enterprise licence above $1M revenue) and Large (API and enterprise self-hosting). Figures (20 May 2026): up to six minutes (Medium 6:20); Small up to two minutes vs 11 s (Open Small) and 47 s (Open); "fully licensed data".Made by Stability AI · 2026-05-20Los Angeles (via Stability AI)stability.ai/news-updates/meet-stable-audio-3-th · stability.ai/news-updateslab's own pages
Stable Audio OpenChips, models and productsAn open-source model for making short audio samples and sound effects. released (open source)Key facts Theme: Speech and audio AI. Announced 5 Jun 2024; research paper post 22 Jul 2024. Stable Audio 3.0 page (20 May 2026) gives its maximum length as 47 seconds.Made by Stability AI · 2024-06-05Read the paperLos Angeles (via Stability AI)stability.ai/news-updates/category/Product · stability.ai/news-updates?offset=1742313654132one source
Stable Audio Open SmallChips, models and productsA small open audio model, released with Arm, that generates sound on phones and other devices. released (open weights)Key facts Theme: Speech and audio AI. Announced 14 May 2025 with Arm (earlier Arm on-device audio post 3 Mar 2025). Stable Audio 3.0 page (20 May 2026) gives its maximum length as 11 seconds.Made by Stability AI · 2025-05-14Los Angeles (via Stability AI)stability.ai/news-updates · stability.ai/news-updates?offset=1742313654132one source
Text to Sound Effects v2 (eleven_text_to_sound_v2)Chips, models and productsElevenLabs model that makes sound effects from a written description. released (API; default model)Key facts Remains the default sound-effects model; SFX v3 (BIZ-LAB-7036) has limited access; access: APIMade by ElevenLabs · — (release date not stated)Londonelevenlabs.io/docs/overview/modelsone source
V2A (video-to-audio research)Chips, models and productsResearch system that generates soundtracks, effects and dialogue to match the action in a video. research (not released)Key facts Announced 17 Jun 2024; not opened to the public pending safety assessments and testing; access: none (research)Made by Google DeepMind · 2024-06-17Londondeepmind.google/blog/generating-audio-for-video/one source
Voice Design (first generative voice model, in Voice Lab)Chips, models and productsElevenLabs model that creates new synthetic voices from scratch rather than copying a real speaker. superseded (product)Key facts Announced 11 Jan 2023; "Design Voice" scheduled for February 2023 inside Voice Lab on the beta platform; successor Voice Design v3 already listed (BIZ-LAB-7027); access: productMade by ElevenLabs · 2023-01-11Londonelevenlabs.io/blog/enter-the-new-year-with-a-banone source
Voice Design v3Chips, models and productsA tool that creates a new synthetic voice from a written description and offers three options. released (product)Key facts Released 25 Jun 2025 in the ElevenLabs dashboard. Models page (read 30 Oct 2026) lists eleven_ttv_v3 for voice design, 70+ languages.Made by ElevenLabs · 2025-06-25Soho, London (W1F 0DQ) (via ElevenLabs)elevenlabs.io/blog/voice-design-v3 · elevenlabs.io/docs/overview/modelslab's own pages
WaveNet (in Google Assistant)Chips, models and productsDeepMind's raw-audio speech generation model (first published 2016), launched in Google Assistant in October 2017. released (in product)Key facts Theme: Speech and audio AI. Original research September 2016 (outside ledger range); launched in Google Assistant October 2017; high-fidelity speech synthesis post November 2017; used to restore voices of speech-impaired users (December 2019).Made by Google DeepMind · 2017-10London (EC2M 4AG) (via Google DeepMind)deepmind.google/blog/page/17/ · deepmind.google/blog/page/14/one source

Tile photo: Neumann U87 Condenser Microphone - Studio A, In Your Ear Studios, Will Fisher from Richmond, VA, United States, CC BY-SA 2.0 (Wikimedia Commons).