{
  "data": null,
  "kind": "anchor.page",
  "links": {
    "api": "https://www.anchorterminal.com/api/v1/index.json",
    "html": "https://www.anchorterminal.com/niches/text-to-speech.voice-agent-turn/",
    "json": "https://www.anchorterminal.com/niches/text-to-speech.voice-agent-turn/index.json",
    "llms": "https://www.anchorterminal.com/llms.txt",
    "markdown": "https://www.anchorterminal.com/niches/text-to-speech.voice-agent-turn/index.md",
    "mcp": "https://www.anchorterminal.com/mcp",
    "slim": "https://www.anchorterminal.com/niches/text-to-speech.voice-agent-turn/index.min.md"
  },
  "markdown": "Two or three companies' docs show a full fit.\n\n| Companies, full fit | With a limit | Carry the tag | Share of the category's work | First-hand accounts |\n| --- | --- | --- | --- | --- |\n| 2 | 19 | 49 | 22 | 15 |\n\n## What counts as done\n\nA connection that lasts a 10-minute call, a cancel message that stops output, and only text sent before the cancel billed.\n\nMeasured, for our own tests: first audio within 300 ms of the first text at the median, audio keeps pace with the text so there is no gap mid-turn, the cancel stops output within one chunk, and no stream silently stops over the call.\n\nHow agents ask for it: say this to the caller as soon as you can; turn my agent's reply into speech on the call; stream the LLM output to a voice and cut it off if the user talks\n\n## Does the job\n\nThe docs show it meets the test above, and nothing we read says otherwise.\n\n- **Deepgram Text-to-Speech (Aura-2, Flux TTS) (Deepgram, Inc.)**: /v2/speak streams Speak tokens into a turn, Flush ends it with SpeechMetadata and a billable_character_count, Interrupt with playback_offset cancels and returns text_spoken, mulaw at 8 kHz is an output, and sessions run an hour; concurrency is 15 in the US and 5 elsewhere.\n- **Inworld Realtime TTS-2 (Theai, Inc. dba Inworld AI)**: WebSocket contexts take LLM fragments with autoModeStrategy SENTENCE_BOUNDARY, flushContext, closeContext and cancelContext; contextClosed is the last message and the billing page says text not yet processed at a cancel is not counted; mu-law and A-law output. acc-13 runs it live and calls it fast.\n\n## Does the job, with a limit\n\nIt does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.\n\n- **Amazon Polly (Amazon Web Services, Inc.)**: StartSpeechSynthesisStream (generative engine only) takes TextEvents as the LLM produces them, with a Force flush and a StreamClosedEvent, but there is no cancel message, the stream closes after 10 minutes or 5 idle seconds, only 8 run at once, and generative voices return no speech marks.\n- **Audio8-TTS-Realistic 0.6B (Edge0 (edge0.ai), which names no legal entity or location on its site. The technical report's authors give their affiliation as AutoArk, and commit metadata in Edge0-AI/Audio8_TTS uses an autoark.ai domain)**: The SGLang Omni adapter serves an OpenAI-compatible /v1/audio/speech with stream true, but it is pinned to one SGLang Omni commit, takes whole text per request with no cancel, and the card says streaming and throughput were not benchmarked for this release.\n- **Azure AI Speech text-to-speech (Microsoft Corporation)**: Input text streaming over the websocket v2 endpoint takes LLM chunks as they arrive, but only through the C#, C++ and Python SDKs, without SSML, with no documented cancel message or billing note, 10 minutes of audio per request, and 429s when a voice's regional capacity is busy.\n- **Cartesia Sonic TTS API + MCP (Cartesia AI, Inc.)**: WebSocket contexts take fragments with continue, max_buffer_delay_ms and a done flag, but cancel only halts requests that have not started generating, Free allows 2 concurrent contexts, acc-29 reports request timeouts cutting calls and acc-30 clipped Japanese endings from collapsed timestamps.\n- **ElevenLabs Text to Speech API + MCP (Eleven Labs Inc.)**: The stream-input WebSocket takes text chunks with chunk_length_schedule, try_trigger_generation, flush and a close message, and the multi-context socket closes one context at a time, but no cancel-billing note exists and acc-04 found real-time latency well above the advertised figure; acc-05, acc-36 and acc-37 run agents on it.\n- **Fish Audio Drama 3 (39 AI, Inc. (the licensor named in the Fish Audio Community License). fish.audio's terms of use name Hanabi AI Inc. as the provider of the hosted services)**: SGLang-Omni serves a streaming /v1/audio/speech with 89 ms median first audio at one stream on an H200 by Fish Audio's own table, but the request takes whole text, cancelling means closing the connection, 48 GB of NVIDIA memory is needed, and acc-31 saw 6 to 20 percent runaway takes.\n- **Fish Audio S2 Pro (39 AI, Inc. (the licensor named in the Fish Audio Research License). fish.audio's terms of use name Hanabi AI Inc. as the provider of the hosted services)**: The api_server streams audio with --streaming and the card claims about 100 ms to first audio on an H200, but the request takes whole text, there is no cancel message, 24 GB of GPU memory is asked for, and the Research License forbids commercial use.\n- **Fish Audio TTS API (Hanabi AI Inc.)**: The WebSocket takes text fragments with flush and stop events and ends with a finish event, but there is no cancel message (stop drains the buffer), billing after a cut is unstated, the 300 ms first audio is Fish Audio's claim, and concurrency is 5 below USD 100 prepaid.\n- **Google Cloud Text-to-Speech (Google LLC)**: StreamingSynthesize takes a stream of text inputs after one config and returns audio as it goes, with MULAW output, but only for Chirp 3 HD voices, with no flush, cancel or done message documented and no note on billing after an early close; billing must be enabled first.\n- **HeyGen Voice (HeyGen Technology, Inc.)**: wss stream-input takes LLM text in pieces with flush, close_context, isFinal and synthesizedChars, and ulaw at 8 kHz is an output option, but 30 requests a minute per member counts every handshake and context, Turbo Free returns 429 on shared capacity, and no latency figure is published.\n- **Ming-3.0-tts-tiny (inclusionAI, Ant Group (named in the site footer of inclusion-ai.org; the LICENSE copyright line reads inclusionAI))**: vLLM-Omni serves /v1/audio/speech with PCM streaming out, but each request needs the whole text, no cancel or done semantics are documented, there are no timestamps, and the recipe targets one H100 or H200.\n- **MiniMax Speech (T2A) (Nanonoble Pte. Ltd.)**: The bidirectional WebSocket takes task_continue at any granularity, buffers to sentences, and has task_flush, task_cancel (keep sending afterwards) and is_final with usage_characters, but T2A is capped at 60 requests a minute, no latency figure is given, and the docs do not say what a cancelled sentence costs.\n- **Murf TTS API + MCP (Murf Inc.)**: The Falcon 2.1 WebSocket takes text in chunks per context_id with end, a clear frame on interruption, a final frame and word timestamps, and streams 8 kHz ULAW; but concurrency is 2 outside US-East (5 there), the 100 ms figure is Murf's own, and billing after clear is unstated.\n- **Qwen3-TTS 12Hz 1.7B CustomVoice (Alibaba Cloud (the copyright holder in the Qwen licences; RDAP names Alibaba Cloud (Singapore) Private Limited as registrant of qwen.ai, from the first batch's check))**: The vLLM-Omni recipe serves /v1/audio/speech with stream true and PCM, but the qwen-tts package has no streaming or server, each request needs the whole text, no cancel exists, and acc-07 builds its own VAD and streaming pipeline around it.\n- **Resemble AI Text-to-Speech API (Resemble AI, Inc.)**: The WebSocket streams audio per request with request_id and a terminal audio_end, but it is Business plan only, each message is a whole text of up to 3,000 characters rather than LLM fragments, and no cancel message exists; the agent must close the socket to stop.\n- **Rime TTS API + MCP (Rime Labs, Inc.)**: Coda WebSocket v1 streams LLM tokens into a context with SPLIT_STRATEGY_ACCUMULATE_SENTENCE, end, stop and cancel, answering cancelled or done, at 8 kHz PCM or mu-law; but Rime publishes no concurrency figure and no rule on what a cancelled context costs, and the 96 ms P50 figure is Rime's own.\n- **Smallest.ai Lightning TTS (Smallest, Inc.)**: WebSocket takes LLM fragments under one context_id, cancel_request discards buffered text, 8 kHz ulaw output and word timestamps exist, but the concurrency page says 1 request per account, no frame marks a context done, and billing after a cancel is not stated.\n- **Soniox Text-to-Speech (Soniox Inc.)**: The WebSocket takes text chunks with text_end, a cancel message stops audio, audio_end and terminated close the stream and 8 kHz output exists, but 3 concurrent streams by default, a stream dies after a few seconds without text, and billing after a cancel is not stated.\n- **VieNeu-TTS v3 Turbo (Phạm Nguyễn Ngọc Bảo, an individual (named on the card as author; Hugging Face account pnnbao-ump, GitHub account pnnbao97). No company found)**: The OpenAI-compatible server streams pcm or wav chunked or over SSE with an 8 kHz resample option, and the author measures 105 to 115 ms first audio on an RTX 3060, but the request takes whole text, cancelling means closing the connection, and it speaks only Vietnamese and English.\n\n- not this job: Ming-3.0-tts-0.5B (inclusionAI (the LICENSE's copyright holder; source headers read Copyright (c) Ant Group)): The publisher's infer.py returns one whole waveform; no streaming, no text input over a socket, no cancel and no server are documented for the 0.5B checkpoint.\n- not this job: OmniVoice (k2-fsa, the Next-gen Kaldi open-source project. The paper's authors are at Xiaomi Corp.; no company is named as the licensor): generate() returns whole waveforms; no streaming, no server endpoint and no cancel in the package, only a Gradio demo and CLI tools.\n\n## What stops agents finishing\n\n- no WebSocket that takes text as it is produced (REST only, or WebSocket only on a Business plan as at Resemble)\n- concurrency of 2 to 5 streams on the plan an agent can sign up to (Cartesia Free 2, Soniox 3, Murf 2 outside US-East)\n- request timeouts with no fallback provider, which leave the caller in silence\n- end of speech only estimated from text, so the call ends before the last word plays\n\n## Around it\n\nBefore: speech-to-text.voice-agent-live, inference.chat-completion. After: voice-agents.play-to-call, telephony.hang-up.\n\n## Where this comes from\n\nWeight 22 of 100 in Text-to-speech: 15 of 46 accounts tagged (acc-03, 04, 05, 06, 07, 11, 13, 18, 28, 29, 30, 36, 37, 41, 44); the job every API vendor leads with and the one with the most failure reports, so weighted at the top\n\nAs data: https://www.anchorterminal.com/api/v1/use-cases/text-to-speech.voice-agent-turn.json, or the MCP tool get_use_case with id text-to-speech.voice-agent-turn. Draft, read 2026-11-12 from public docs and first-hand accounts, not yet reviewed by a person.\n",
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "attributionRequired": true,
    "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-11-16",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.8",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run",
    "terms": "https://www.anchorterminal.com/terms/#data"
  },
  "page": {
    "breadcrumbs": [
      {
        "name": "Home",
        "url": "https://www.anchorterminal.com/"
      },
      {
        "name": "For agents",
        "url": "https://www.anchorterminal.com/agents/"
      },
      {
        "name": "Niches",
        "url": "https://www.anchorterminal.com/niches/"
      },
      {
        "name": "Speak an LLM reply to a live caller as it streams, and stop when interrupted",
        "url": ""
      }
    ],
    "description": "Speak an LLM reply to a live caller as it streams, and stop when interrupted: 2 companies fully do it, 19 with a limit, 49 carry the tag. Every listing in Text-to-speech with the reason and a source.",
    "facts": null,
    "h1": "Speak an LLM reply to a live caller as it streams, and stop when interrupted",
    "image": "https://www.anchorterminal.com/assets/og/default.png",
    "path": "/niches/text-to-speech.voice-agent-turn/",
    "published": "",
    "section": "agents",
    "title": "Speak an LLM reply to a live caller as it streams, and stop when…",
    "toc": null,
    "updated": "2026-11-15",
    "url": "https://www.anchorterminal.com/niches/text-to-speech.voice-agent-turn/"
  },
  "tokens": {
    "markdown": 2800,
    "slim": 630
  },
  "version": 1
}
