Audio Generation

Mesh API provides a full suite of audio endpoints — convert text to speech, transcribe audio files, stream TTS/STT in real time, and browse available voices — all through a single API key.

All endpoints share the same base URL: https://api.meshapi.ai/v1/audio

Auth: Authorization: Bearer rsk_<your-key> on all REST requests. WebSocket endpoints accept the key via Sec-WebSocket-Protocol: Bearer <rsk_...> or ?api_key=<rsk_...>.

Speech-to-Text

POST /v1/audio/transcriptions

Transcribe an audio file. The provider is resolved automatically from the model name. You can supply audio as a file upload, a public URL, or a cloud storage URL.

This endpoint uses multipart/form-data — not JSON.

Multiple STT providers are supported — e.g. elevenlabs/scribe_v1 (default) and elevenlabs/scribe_v2, and sarvam/saaras:v2. To list transcription-capable models, use the search endpoint filtered by modality: GET /v1/models/search?input_modality=audio&output_modality=text.

Form fields

Response

1
{ "text": "The transcribed text goes here." }

Examples

$
curl -X POST https://api.meshapi.ai/v1/audio/transcriptions \
>
  -H "Authorization: Bearer rsk_..." \
>
  -F "model=elevenlabs/scribe_v2" \
>
  -F "file=@recording.mp3" \
>
  -F "language_code=en-US" \
>
  -F "diarize=true"

Transcribe and Translate

POST /v1/audio/transcriptions/translate

Transcribe audio and translate the result directly to English in a single step. Uses Sarvam models by default.

This endpoint uses multipart/form-data.

Form fields

Response

1
{ "text": "The English translation of the transcribed audio." }

Example

$
curl -X POST https://api.meshapi.ai/v1/audio/transcriptions/translate \
>
  -H "Authorization: Bearer rsk_..." \
>
  -F "model=sarvam/saaras:v2" \
>
  -F "file=@hindi_recording.wav"

WebSocket Real-Time STT

WS /v1/audio/transcriptions/realtime

Transcribe audio in real time. Send raw audio chunks as they are captured (e.g. from a microphone) and receive partial and final transcripts as they are produced.

The wire protocol is selected automatically from the model you pass. Standard streaming models (e.g. openai/whisper-large-v3) use an OpenAI-compatible frame protocol. ElevenLabs models (elevenlabs/scribe_v2_realtime) use ElevenLabs’ Scribe v2 realtime frames — see the ElevenLabs models subsection below.

Authentication

Pass your Mesh API key in one of these ways:

• Sec-WebSocket-Protocol: Bearer rsk_... header
• ?api_key=rsk_... query parameter
• ?token=rsk_... query parameter

Query parameters

Supported audio formats: pcm_8000, pcm_16000, pcm_22050, pcm_24000, pcm_44100, pcm_48000, ulaw_8000

Standard streaming models

Standard streaming models — e.g. openai/whisper-large-v3 — use an OpenAI-compatible frame protocol.

Client → server (JSON frames)

1
{
2
  "type": "input_audio_buffer.append",
3
  "audio": "<base64-encoded PCM s16le 16 kHz mono>"
4
}

Server → client (JSON frames)

Example

1
const ws = new WebSocket(
2
  "wss://api.meshapi.ai/v1/audio/transcriptions/realtime" +
3
    "?model=openai/whisper-large-v3&api_key=rsk_..."
4
);
5
6
ws.onopen = () => {
7
  console.log("Connected. Start sending audio chunks.");
8
};
9
10
ws.onmessage = ({ data }) => {
11
  const frame = JSON.parse(data);
12
  if (frame.type === "conversation.item.input_audio_transcription.delta") {
13
    process.stdout.write("\r" + frame.delta);
14
  } else if (
15
    frame.type === "conversation.item.input_audio_transcription.completed"
16
  ) {
17
    console.log("\nFinal:", frame.transcript);
18
  }
19
};
20
21
// Send a PCM audio chunk (s16le, 16 kHz, mono — from a microphone, for example)
22
function sendAudioChunk(pcmBuffer) {
23
  const b64 = Buffer.from(pcmBuffer).toString("base64");
24
  ws.send(
25
    JSON.stringify({
26
      type: "input_audio_buffer.append",
27
      audio: b64,
28
    })
29
  );
30
}
31
32
// Commit the buffered audio to get a final transcript
33
function commit() {
34
  ws.send(JSON.stringify({ type: "input_audio_buffer.commit" }));
35
}

ElevenLabs models

ElevenLabs models (elevenlabs/scribe_v2_realtime) use ElevenLabs’ Scribe v2 realtime frames.

Client → server (JSON frames)

Send input_audio_chunk frames with base64-encoded audio. This is the only message type the server forwards upstream — any other message type is silently dropped.

1
{
2
  "message_type": "input_audio_chunk",
3
  "audio_base_64": "<base64-encoded PCM audio>",
4
  "sample_rate": 16000,
5
  "commit": false
6
}

Set "commit": true to trigger a VAD commit when using commit_strategy: manual.

Server → client (JSON frames)

1
const ws = new WebSocket(
2
  "wss://api.meshapi.ai/v1/audio/transcriptions/realtime" +
3
    "?model=elevenlabs/scribe_v2_realtime&audio_format=pcm_16000&commit_strategy=manual&api_key=rsk_..."
4
);
5
6
ws.onopen = () => {
7
  console.log("Connected. Start sending audio chunks.");
8
};
9
10
ws.onmessage = ({ data }) => {
11
  const frame = JSON.parse(data);
12
  if (frame.message_type === "partial_transcript") {
13
    process.stdout.write("\r" + frame.text);
14
  } else if (frame.message_type === "committed_transcript") {
15
    console.log("\nFinal:", frame.text);
16
  }
17
};
18
19
// Send a PCM audio chunk (from a microphone, for example)
20
function sendAudioChunk(pcmBuffer) {
21
  const b64 = Buffer.from(pcmBuffer).toString("base64");
22
  ws.send(
23
    JSON.stringify({
24
      message_type: "input_audio_chunk",
25
      audio_base_64: b64,
26
      sample_rate: 16000,
27
    })
28
  );
29
}
30
31
// Commit the buffered audio to get a final transcript
32
function commit() {
33
  ws.send(
34
    JSON.stringify({
35
      message_type: "input_audio_chunk",
36
      audio_base_64: "",
37
      sample_rate: 16000,
38
      commit: true,
39
    })
40
  );
41
}

Voice Management

List voices

GET /v1/audio/voices

Returns a unified voice catalog spanning every TTS model brand. The list can be large, so filter by brand or model to narrow it to the voices you can actually use with the model you plan to call.

Each voice is returned with voice_id, name, brand, provider, model, language, gender, and preview_url. The response also carries count and a truncated flag (true when the list was capped before the full set was fetched).

$
# All voices for a given brand
$
curl "https://api.meshapi.ai/v1/audio/voices?brand=elevenlabs&search=rachel" \
>
  -H "Authorization: Bearer rsk_..."
$
$
# Voices usable with a specific model
$
curl "https://api.meshapi.ai/v1/audio/voices?model=hexgrad/kokoro-82m" \
>
  -H "Authorization: Bearer rsk_..."

Get a single voice

GET /v1/audio/voices/{voice_id}

Fetch metadata for a specific voice by its ElevenLabs voice ID.

$
curl https://api.meshapi.ai/v1/audio/voices/JBFqnCBsd6RMkjVDRZzb \
>
  -H "Authorization: Bearer rsk_..."

Error handling

All endpoints use standard HTTP status codes. Common cases:

WebSocket sessions send a JSON error frame and then close with code 1000 before disconnecting.