Realtime TTS-2 is live. Built for realtime conversation that feels human. Learn more

TextToSpeech

Synthesize speech (stream)

Receive audio chunks as newline-delimited JSON (NDJSON), with one JSON object per line. Successful messages wrap their payload in result. When present, result.audioContent contains Base64-encoded audio; messages may also contain result.usage and result.timestampInfo. Timestamp-only messages may follow the audio. HTTP read boundaries are not message boundaries. Check the HTTP status and any error object in the stream. See the HTTP streaming guide for parsing, audio framing, usage, and cancellation.

POST/tts/v1/voice:stream

HTTP request and response framing

Send POST https://api.inworld.ai/tts/v1/voice:stream with Authorization: Basic <INWORLD_API_KEY> and Content-Type: application/json. Use the Portal's Base64 credentials unchanged, without encoding them again. Keep the key on your server; see Authentication for browser and mobile token options.

bash
curl --no-buffer 'https://api.inworld.ai/tts/v1/voice:stream' \
  --header "Authorization: Basic $INWORLD_API_KEY" \
  --header 'Content-Type: application/json' \
  --header 'Accept: application/json' \
  --data '{
    "text": "Hello, world!",
    "voiceId": "Ashley",
    "modelId": "inworld-tts-2",
    "audioConfig": { "audioEncoding": "PCM", "sampleRateHertz": 48000 }
  }'

The JSON response is newline-delimited JSON (NDJSON): one JSON object per line, not SSE (data: events) or a single JSON document. Each successful message wraps its payload in result. Audio is Base64-encoded inside result.audioContent, not raw audio in the HTTP body. Example lines below use placeholders for the audio bytes:

json
{"result":{"audioContent":"<base64-audio>","usage":{"processedCharactersCount":13,"modelId":"inworld-tts-2"}}}
{"result":{"audioContent":"<base64-audio>","usage":{"processedCharactersCount":0,"modelId":"inworld-tts-2"}}}

A network read may contain part of a line or several lines. Buffer until a newline, parse each complete object, and Base64-decode its audio independently. Do not call response.json() on the whole streaming response or concatenate Base64 strings before decoding. Process a final non-empty line at EOF too. Timestamp-only messages can arrive after the audio; keep reading until EOF.

Request fieldMeaning
textUp to 4,000 UTF-16 code units per HTTP streaming request
voiceIdUse the exact voiceId from List voices, such as Ashley
modelIdSet explicitly; choose from Models, for example inworld-tts-2
audioConfig.audioEncodingEncoding of the decoded audio bytes; see the table below
audioConfig.sampleRateHertzRequested output sample rate in Hz; configure your PCM player to match

The non-streaming endpoint accepts 2,000 UTF-16 code units, while this endpoint accepts 4,000. In JavaScript, text.length measures this unit; some emoji count as two. An oversized request returns INVALID_ARGUMENT (code 3) with a message such as text length should not exceed 4000 characters. SDK versions can enforce their own smaller limits or split text into multiple requests. Set modelId explicitly and consult the documentation for the installed SDK version instead of relying on its default model or chunk size.

Choose an audio encoding

The table below compares common streaming choices; it is not the full encoding list. The API also exposes OGG_OPUS, ALAW, MULAW, and FLAC. See the audioConfig.audioEncoding API reference for the complete enum and format descriptions.

EncodingDecoded bytes in an HTTP stream
PCMRaw signed 16-bit little-endian samples, with no WAV header; suitable for a PCM playback queue
LINEAR16A complete WAV header in every audio chunk; do not feed the headers to a raw PCM player or concatenate chunks as one ordinary WAV file
WAVA streaming WAV header in the first audio chunk only
MP3Compressed audio; use an MP3-capable streaming decoder/player

The encoding names describe the audio after Base64 decoding. PCM and LINEAR16 contain the same sample format but have different framing.

Read the stream in Node.js

This server-side example uses Node.js 20+ built-in fetch. onAudio receives decoded PCM bytes; enqueue them for playback or write them to a raw PCM file. A raw PCM file needs its sample rate and channel count supplied to the player.

javascript
async function streamSpeech(text, onAudio, { signal } = {}) {
  const response = await fetch('https://api.inworld.ai/tts/v1/voice:stream', {
    method: 'POST',
    headers: {
      Authorization: `Basic ${process.env.INWORLD_API_KEY}`,
      'Content-Type': 'application/json',
      Accept: 'application/json',
    },
    body: JSON.stringify({
      text,
      voiceId: 'Ashley',
      modelId: 'inworld-tts-2',
      audioConfig: { audioEncoding: 'PCM', sampleRateHertz: 48000 },
    }),
    signal,
  });
  if (!response.ok) {
    const body = await response.text();
    let message = body;
    try {
      const payload = JSON.parse(body);
      message = (payload.error ?? payload).message ?? body;
    } catch { /* An intermediary may return a non-JSON error. */ }
    throw new Error(`HTTP ${response.status}: ${message}`);
  }
  if (!response.body) throw new Error('Missing response stream');

  const decoder = new TextDecoder();
  let pending = '';
  let processedCharacters = 0;
  async function consume(line) {
    if (!line.trim()) return;
    const message = JSON.parse(line);
    if (message.error) throw new Error(message.error.message ?? 'Synthesis failed');
    const result = message.result;
    processedCharacters += result?.usage?.processedCharactersCount ?? 0;
    if (result?.audioContent) {
      await onAudio(Buffer.from(result.audioContent, 'base64'));
    }
  }

  for await (const bytes of response.body) {
    pending += decoder.decode(bytes, { stream: true });
    let newline;
    while ((newline = pending.indexOf('\n')) !== -1) {
      const line = pending.slice(0, newline);
      pending = pending.slice(newline + 1);
      await consume(line);
    }
  }
  pending += decoder.decode();
  await consume(pending);
  return { processedCharacters };
}

Errors, usage, and cancellation

Check both the initial HTTP status and each parsed message. A stream can return an HTTP 200 and later end with an error message:

json
{"error":{"code":3,"message":"Invalid request"}}

Non-streaming endpoints can instead return {"code":5,"message":"Voice not found"}. For shared error handling, inspect payload.error ?? payload. Codes in these objects are gRPC status codes, not HTTP status numbers. Preserve any audio already delivered as partial output and report the failure; a successful HTTP status alone does not establish that synthesis finished.

For HTTP streaming, usage.processedCharactersCount carries the full input's UTF-16 length on the first response message; subsequent messages report zero. Sum this field across messages, or retain the first value. Reading only the last message reports zero and loses the request's usage.

Pass an AbortController signal to streamSpeech and call abort() to cancel the fetch and response reader. Handle the rejection as cancellation, and stop and clear your playback queue separately. Cancellation does not refund text already metered: the full submitted text is metered when the first response is emitted, rather than in proportion to the audio you play. Aborting before you receive audio is not proof that the server has not already metered the request.

Audio chunk size, arrival spacing, and time to first audio vary with text, model, encoding, and load. Buffer for continuous playback rather than assuming fixed chunk durations or an arrival rate equal to playback speed. For mono PCM16, durationSeconds = decodedByteLength / (2 * sampleRateHertz); at 48 kHz, 9,600 decoded bytes represent 100 ms of audio. These are sizing calculations, not latency guarantees.

Authorizations

Authorizationstringrequired

Your authentication credentials. For Basic authentication, please populate Basic $INWORLD_API_KEY. You can create a key in one command with the Inworld CLI: inworld workspace add-key.

Body

application/json

textstringrequired

The text to be synthesized into speech. Maximum input of 4,000 UTF-16 code units (JavaScript text.length) per HTTP streaming request. Exceeding the limit returns INVALID_ARGUMENT (code 3). SDK versions may apply smaller limits or split the input into separate requests.

voiceIdstring

The ID of the voice to use for synthesizing speech. Set either voiceId or voiceDesign, not both.

audioConfigobject

Configurations to use when synthesizing speech.

Show child attributes

audioEncodingenum<string>default: "MP3"

The desired output format of the synthesized audio. Defaults to MP3.

  • LINEAR16: Uncompressed 16-bit signed little-endian samples (Linear PCM). For non-streaming, the WAV header is included in the response. For streaming, the WAV header is included in every audio chunk.
  • MP3: MP3 audio.
  • OGG_OPUS: Opus encoded audio wrapped in an ogg container. The result will be a file which can be played natively on Android, and in browsers (at least Chrome and Firefox). The quality of the encoding is considerably higher than MP3 while using approximately the same bitrate.
  • ALAW: ALAW encoded audio. 8-bit companded PCM.
  • MULAW: MULAW encoded audio. 8-bit companded PCM.
  • FLAC: FLAC encoded audio. Lossless audio format.
  • PCM: PCM audio. Uncompressed 16-bit signed little-endian samples with no WAV header.
  • WAV: WAV audio. Uncompressed 16-bit signed little-endian samples. For non-streaming, the WAV header is included in the response. For streaming, the WAV header is included in the first audio chunk only.

Available options:LINEAR16MP3OGG_OPUSALAWMULAWFLACPCMWAV

bitRateinteger

Bits per second of the audio. Only for compressed audio formats (MP3, OGG_OPUS). The default is 128,000.

sampleRateHertzinteger

The synthesis sample rate (in hertz) for this audio. Accepts values within the range [8000, 48000]. Supported sample rates are: 8000, 16000, 22050, 24000, 32000, 44100, 48000.

When this is specified, if this is different from the voice's natural sample rate, then the audio will be converted to the desired sample rate (which might result in worse audio quality), unless the specified sample rate is not supported for the encoding chosen, in which case it will fail the request and return an error. The default is 48,000.

speakingRatenumber

Speaking rate/speed, in the range [0.5, 1.5]. The default is 1.0, which is the normal native speed supported by the specific voice. We recommend using values above 0.8 to ensure high quality.

modelIdstringrequired

The ID of the model to use for synthesizing speech. See Models for available models.

languagestring

BCP-47 language tag (e.g., en-US, fr-FR, ja-JP) specifying the language that the given voice should speak the text in. Matching is case- and separator-insensitive for standard two-part tags (en-gb, EN_GB, and en-GB are equivalent); longer tags with extension subtags must match a catalog entry exactly. If a localized voice prompt exists for the language, it will be used. When omitted, the original voice prompt will be used and the language will be auto-detected from the input text. If an invalid language code is provided, an error will be returned.

See Languages for more details.

deliveryModeenum<string>default: "DELIVERY_MODE_UNSPECIFIED"

Only supported by `inworld-tts-2`. The field is ignored on other models.

Controls how varied the output is.

  • DELIVERY_MODE_UNSPECIFIED: Defaults to BALANCED behavior.
  • STABLE: Optimizes for more consistent, predictable output.
  • BALANCED: Balanced between stability and diversity.
  • CREATIVE: Optimizes for increased emotional range and variation.

Available options:DELIVERY_MODE_UNSPECIFIEDSTABLEBALANCEDCREATIVE

instructionstring

Only supported by `inworld-tts-2`. The field is ignored on other models.

Speaking-style instruction for this request — for example speak loudly and urgently or sound out of breath. Applies to the whole request. Write it in English, even when text is in another language. An empty string means unset.

You can also change the instruction mid-text with inline [bracket] tags. A tag applies from where it appears until you change it, so it overrides this field from that point on; [reset] removes the instruction for the rest of the text. Prefer one approach or the other rather than combining them.

See Steering for the full guide.

temperaturenumberdefault: 1

Ignored on `inworld-tts-2`. Use [`deliveryMode`](#body-delivery-mode) instead.

Determines the degree of randomness when sampling audio tokens to generate the response.

Defaults to 1.0. Accepts values between 0 (exclusive) and 2 (inclusive). Higher values will make the output more random and can lead to more expressive results. Lower values will make it more deterministic. If 0 is provided, the default value will be used.

For the most stable results, we recommend using the default value.

timestampTypeenum<string>default: "TIMESTAMP_TYPE_UNSPECIFIED"

Controls timestamp metadata returned with the audio. When enabled, the response includes timing arrays, which can be useful for word-highlighting, karaoke-style captions, and lipsync.

  • WORD: Output arrays under timestampInfo.wordAlignment (words, wordStartTimeSeconds, wordEndTimeSeconds).
  • CHARACTER: Output arrays under timestampInfo.characterAlignment (characters, characterStartTimeSeconds, characterEndTimeSeconds).
  • TIMESTAMPTYPEUNSPECIFIED: Do not compute alignment; timestamp arrays will be empty or omitted.

Phonetic details: phoneticDetails is currently only returned for WORD alignment (not CHARACTER).

Latency note: Alignment adds additional computation. Enabling alignment can increase latency.

Available options:TIMESTAMP_TYPE_UNSPECIFIEDWORDCHARACTER

applyTextNormalizationenum<string>default: "APPLY_TEXT_NORMALIZATION_UNSPECIFIED"

When enabled, text normalization automatically expands and standardizes things like numbers, dates, times, and abbreviations before converting them to speech. For example, Dr. Smith becomes Doctor Smith, and 3/10/25 is spoken as March tenth, twenty twenty-five. Turning this off may reduce latency, but the speech output will read the text exactly as written. Defaults to automatically deciding whether to apply text normalization.

Available options:APPLY_TEXT_NORMALIZATION_UNSPECIFIEDONOFF

enhanceGenerationbooleandefault: false

When true, applies denoising to the synthesized audio to reduce background noise and artifacts, improving the overall audio quality of the generation. Defaults to false (no denoising).

synthesisContextobject

Context for the current synthesis request. Supplying the text of earlier requests from the same session or conversation gives the model additional context and can improve the quality of the generation, especially for short or ambiguous input text. Context text is not billed. The texts of all previous requests combined must not exceed 2,000 characters.

Show child attributes

previousRequestsobject[]

Previous requests from the same session or conversation, in the order they were synthesized.

Show child attributes

textstring

The text that was synthesized in the previous request.

voiceDesignobject

Preview. Design a voice for this synthesis request without creating or saving a voice. Supported by inworld-tts-2 on unary, server-streaming, async, and batch synthesis. Omit voiceId. Each request or batch item designs its own voice; repeating a description does not guarantee the same voice. See Ad-hoc voice design.

Show child attributes

designPromptstringrequired

A description of the desired voice, such as its accent, pitch, timbre, and delivery. Use 7–1,024 characters; leading and trailing whitespace does not count toward the minimum. This describes the voice; put the words to speak in text.

timestampTransportStrategyenum<string>default: "TIMESTAMP_TRANSPORT_STRATEGY_UNSPECIFIED"

The transport strategy of timestamps info.

  • TIMESTAMP_TRANSPORT_STRATEGY_UNSPECIFIED: The service will automatically decide the transport strategy.
  • SYNC: Timestamps will be returned in the same message as the audio data.
  • ASYNC: Timestamps could return in trailing message after the audio data. Use this strategy to reduce latency of the first audio chunk.

Available options:TIMESTAMP_TRANSPORT_STRATEGY_UNSPECIFIEDSYNCASYNC

Response

200 - application/json

resultobject

A chunk containing audio in the requested format. PCM chunks are raw and headerless. LINEAR16 chunks each contain a complete WAV header. For WAV, only the first chunk contains the header.

Show child attributes

audioContentstring

Base64-encoded audio bytes in the requested encoding. PCM is headerless signed 16-bit little-endian audio. LINEAR16 includes a WAV header in the non-streaming response and in every HTTP streaming audio chunk; WAV includes a header only in the first streaming audio chunk. Decode each audioContent value separately before playback or concatenation.

Maximum output audio size of 16MB. To avoid errors with longer texts, please use a compressed audio format with an appropriate bit rate, or use the streaming endpoint.

usageobject

Usage information for the request.

Show child attributes

processedCharactersCountinteger

Number of input UTF-16 code units processed. For HTTP streaming, the first response message carries the full count and subsequent messages report zero. Sum across messages rather than reading only the last message. Cancelling after the request is metered does not reduce the count to the audio played.

modelIdstring

The model used for speech synthesis.

timestampInfoobject

Timestamp alignment information (present when alignment is enabled).

Show child attributes

wordAlignmentobject

Word-level alignment when timestampType is WORD.

Show child attributes

wordsstring[]

Aligned words in order.

wordStartTimeSecondsnumber[]

Start time for each word in seconds from the beginning of the audio.

wordEndTimeSecondsnumber[]

End time for each word in seconds from the beginning of the audio.

phoneticDetailsobject[]

Detailed phoneme-level timing and viseme information, useful for precise lip-sync animation.

Show child attributes

wordIndexinteger

Index of the word this phonetic detail belongs to (0-based).

phonesobject[]

Array of phonemes that make up this word.

Show child attributes

phoneSymbolstring

The phoneme symbol (IPA notation).

startTimeSecondsnumber

Start time of the phoneme in seconds.

durationSecondsnumber

Duration of the phoneme in seconds.

visemeSymbolstring

The viseme symbol for lip-sync animation (e.g., aei, o, bmp, fv, l, r, th, qw, ee, chjsh, cdgknstxyz).

isPartialboolean

True when the server considers the word potentially unstable (e.g., last word in a non-final streaming update). Clients may choose to delay processing partial words until isPartial becomes false.

characterAlignmentobject

Character-level alignment when timestampType is CHARACTER.

Show child attributes

charactersstring[]

Aligned characters (including punctuation and spaces) in order.

characterStartTimeSecondsnumber[]

Start time for each character in seconds from the beginning of the audio.

characterEndTimeSecondsnumber[]

End time for each character in seconds from the beginning of the audio.

errorobject

A response may contain an error object if an error happens in the stream.

Show child attributes

codeinteger

The error code, as specified by gRPC status codes

messagestring

A short description of the error.

detailsobject[]

Show child attributes

@typestring