FAQ - Modulate

Authentication and API keys

How do I get an API key? Create a free account and your API key will be available in the dashboard after sign-up. How do I authenticate my requests? Authentication works differently depending on the API type. For REST endpoints, pass your key in the X-API-Key header:

X-API-Key: your_api_key_here

For WebSocket endpoints, pass it as a query parameter at connection time:

wss://platform.modulate.ai/api/velma-2-stt-streaming?api_key=your_api_key_here

Is it safe to put my API key in code?

Never commit credentials to source control. API keys pushed to a repository — even briefly — should be considered compromised. Rotate the key immediately if this happens.

Store your key in a .env file, load it at runtime via python-dotenv or your environment’s secret manager, and ensure .env is in your .gitignore. The Quick start guide covers this setup in full.

Models and capabilities

What models are available?

Model	API type	Primary use
Multilingual Transcription Batch	REST	Transcription with the full feature set
Multilingual Transcription Streaming	WebSocket	Real-time transcription with per-utterance results
English Fast Transcription Batch	REST	High-throughput English transcription at lowest cost
English Fast Transcription Streaming	WebSocket	Low-latency English transcription with partials every ~1.5s
Multilingual Fast Transcription Batch	REST	Fast multilingual transcription with optional language declaration
Multilingual Fast Transcription Streaming	WebSocket	Low-latency multilingual transcription across 25 auto-identified languages
Deepfake Detection Batch	REST	Analyzing recorded audio files for synthetic speech
Deepfake Detection Streaming	WebSocket	Live deepfake detection with results from 500ms onward
Emotion Detection Batch	REST	Classify the emotional tone of an audio file
Accent Detection Batch	REST	Classify the speaker accent of an audio file
PII/PHI Redaction Batch	REST	Transcription with PII/PHI text redaction and audio silencing
PII/PHI Redaction Streaming	WebSocket	Real-time PII/PHI redaction — redacted transcript and MP3 clips per utterance
Music & Speech Detection Batch	REST	Classify audio as music, speech, or neither
Music & Speech Detection Streaming	WebSocket	Real-time audio classification as music, speech or neither
AI Music Detection Batch	REST	Detect AI-generated music in audio
AI Music Detection Streaming	WebSocket	Real-time detection of AI-generated music in audio
Audio Event Detection Batch	REST	Detect which of 42 non-speech sound events are present in an audio file
Language Detection	REST	Identify the spoken language of an audio file

What languages does transcription support? The Multilingual Transcription batch and streaming models support 100 languages with automatic language detection. Language is detected per-utterance, so code-switching within a single file is handled automatically. Multilingual Fast Transcription (batch) also transcribes any supported language — declare it with the optional language parameter or let it be detected automatically. Multilingual Fast Transcription (streaming) covers 25 languages, always auto-identified, and reports each utterance’s language when detect_language=true. The English Fast Transcription model is English-only. What optional enrichments are available on transcription? The Multilingual Transcription batch and streaming models support the following optional fields, configurable per request:

Parameter	What it adds	Default
`speaker_diarization`	Distinct speaker labels per utterance	`true`
`emotion_signal`	Detected emotional tone (e.g. Neutral, Happy, Frustrated)	`false`
`accent_signal`	Detected accent (e.g. American, British, Indian)	`false`
`deepfake_signal`	Deepfake score per utterance (0.0 = human, 1.0 = synthetic)	`false`
`pii_phi_tagging`	PII/PHI wrapped with tags in transcript text	`false`

English Fast Transcription (batch) supports two independent opt-in flags, both defaulting to false, and none of the other enrichments:

Parameter	What it adds	Default
`time_stamps`	A `words` array giving each word a start time, end time, and alignment confidence	`false`
`speaker_diarization`	An `utterances` array of speaker-labeled turns with text and timing	`false`

words times are in seconds, while utterances times are in milliseconds; both are measured from the start of the file. See English Fast Transcription (batch) for the full response shape. English Fast Transcription (streaming) has one optional enrichment of its own:

Parameter	What it adds	Default
`diarize`	`speaker`, `n_speakers`, `speakers_capped`, and `speaker_purity` on each final `utterance`	`false`

The speaker index is stable only within one connection, so it cannot be matched across calls. See English Fast Transcription (streaming). What’s the difference between PII/PHI tagging and PII/PHI redaction? These are two separate capabilities with different outputs:

PII/PHI tagging (pii_phi_tagging=true on Multilingual Transcription batch or streaming) — identifies PII/PHI spans in the transcript and wraps them with tags. The original text content is preserved. Use this when downstream systems need to detect or handle sensitive spans while retaining the full transcript.
PII/PHI redaction (PII/PHI Redaction APIs) — replaces each detected PII/PHI span with an empty marker tag (e.g. <pii:name></pii:name>, <pii:ssn></pii:ssn>, <phi></phi>) in the transcript and silences the corresponding audio ranges in the returned MP3. Use this when the audio itself must be clean — for example, recordings that will be shared, archived, or reviewed by parties who should not hear sensitive information.

What types of PII/PHI get redacted? Currently, every entity type the model can detect is redacted, covering both personal information (PII) and health information (PHI). For more detail, see the PII/PHI Redaction Batch reference in the API Reference tab. Can I configure which PII/PHI tags are enabled? Not at this time, but it’s on our roadmap. Reach out to Support to let us know what your specific needs are. How accurate is the deepfake detection? The Deepfake Detection model scores 1.104% average EER across the 14 datasets of the Hugging Face Speech DF Arena benchmark, ranked 1st as of 11 March 2026. Quoted as accuracy that is the same figure restated: 98.9%. Deepfake Detection results carries the competitor comparison, both EER metrics, and the live source link. What does the deepfake confidence score mean? Confidence represents how certain the model is in its verdict, on a scale of 0 to 1. A frame with verdict: "synthetic" and confidence: 0.97 means the model is highly confident that segment contains AI-generated speech. A no-content verdict indicates the frame is silent or contains no usable audio — these frames are not sent through the model and always return confidence: 1.0.

Audio formats and file requirements

What audio formats are supported? The accepted set differs per endpoint, and each capability page lists its own in full. Deepfake Detection, Music & Speech Detection, AI Music Detection, Language Detection, and English Fast Transcription accept 96 file extensions in batch. All of those except Language Detection, which is batch-only, accept 79 container formats plus 22 raw formats in streaming. The remaining batch endpoints accept AAC, AIFF, FLAC, MOV, MP3, MP4, OGG, Opus, WAV, and WebM, and the remaining streaming endpoints accept 7 container formats plus 20 raw formats. Streaming endpoints split into two groups. Transcription, Redaction, and Velma Triage auto-detect self-describing containers, so audio_format is optional. Deepfake Detection, Music & Speech Detection, AI Music Detection, and English Fast Transcription require audio_format on every connection, including containers, and omitting it closes the connection with 1003. Raw PCM, mu-law, and A-law formats are headerless everywhere they are accepted, so they always require sample_rate and num_channels alongside audio_format. Is there a file size limit? 100 MB for all batch endpoints. Is there a minimum audio length? For deepfake detection, audio must be at least 0.5 seconds. Files shorter than this are rejected with a 422. For transcription, very short clips may return empty or minimal results. What is the recommended audio length for deepfake detection? 4–60 seconds is the recommended range. Files shorter than one full 4-second analysis window are padded before inference. Leading and trailing silence is trimmed automatically — frame timestamps reflect positions in the original file.

Pricing and billing

How is usage billed? Billing is credit-based and priced per hour of audio processed. Prices range from $0.01/hour to $1.25/hour depending on the model. Review pricing for all models here: platform.modulate.ai/pricing Is there a free tier? Yes. Free credits are included when you create an account — no credit card required to get started. Where can I monitor my usage? Real-time usage and billing details are in the Usage dashboard.

Rate limits

What rate limits apply? Concurrency is capped per model: the number of simultaneous in-flight requests or active WebSocket connections against one endpoint. The default is 3. Credits are tracked separately and are not a rate limit. What happens when I hit the concurrency cap? REST endpoints return 429. WebSocket connections close with code 4030. The Usage dashboard shows current usage.

Production recommendation: bound parallel requests with a semaphore sized to the cap. Retrying into a full queue does not help. Caps are per model, so spreading load across models raises total throughput.

Can the cap be increased? Contact support@modulate.ai with the model and expected traffic.

Streaming (WebSocket)

How do I signal the end of my audio stream? Send an empty text frame ("") on the open connection. The server will drain any buffered audio, deliver outstanding results, send a done message, then close the connection cleanly. What does the done message look like?

Multilingual Transcription streaming done

{
  "type": "done",
  "duration_ms": 45000
}

Deepfake streaming done

{
  "type": "done",
  "duration_ms": 12500,
  "frame_count": 10
}

What are partial results in Multilingual Transcription streaming? When partial_results=true, the server emits partial_utterance messages while speech is in progress, before the utterance is finalized. Each partial replaces the previous one — the final utterance message supersedes all preceding partials for that segment. With the deepfake signal enabled, a partial may also carry an interim deepfake_score (null until a value is available). Emotion and accent labels arrive on the finalized utterance only. Useful for live caption rendering where low perceived latency matters. What WebSocket close codes should I handle?

WebSocket close code reference

Code	Meaning
`1000`	Normal closure — received after `done`, connection finished cleanly
`1003`	Invalid query parameters (bad `audio_format`, `sample_rate`, or `num_channels`), or a protocol/configuration error
`1011`	Internal server error, or processing ended before completion
`4001`	Invalid API key
`4002`	Audio could not be decoded or doesn’t match the declared format
`4003`	The request is not permitted
`4004`	The API key does not have access to this model
`4029`	Insufficient credits
`4030`	Concurrent request limit reached
`4031`	Monthly usage limit reached

Errors

What does a 503 response mean? The inference server is temporarily overloaded. Wait a moment and retry. For production workloads, implement exponential backoff with jitter rather than an immediate retry loop. What does a 504 response mean? The request timed out — batch processing has a 60-second limit. This is uncommon for typical audio lengths. If you see it consistently on files within the recommended size range, contact support@modulate.ai. My file was rejected with 422. Why? The audio is too short for analysis. Deepfake detection requires a minimum of 0.5 seconds. Check the actual duration of your file — empty or near-silent files sometimes report a longer duration than their usable content.

Privacy and data

Does Modulate sell the audio I send through the API? No. Modulate does not sell personal data, including audio submitted through the API. Audio processed via the platform is used to deliver the service and to improve Modulate’s models. See the training question below for details. Is my audio used to train Modulate’s AI models? It depends on your account type:

Self-service (pay-as-you-go) customers are automatically enrolled in the Model Improvement Program as a condition of using the platform. An opt-out option is available by contacting sales@modulate.ai. Audio you submit may be used to train and improve Modulate’s models.
Enterprise customers (annual commitment): whether you participate in the Model Improvement Program depends on your service tier and is confirmed in your Order Form. Participation may affect your per-credit rate. Contact your account representative or sales@modulate.ai to discuss options.

Opting out does not affect the quality or functionality of your results. Separately from training, Modulate may review a sample of submitted audio to offer support, troubleshoot customer issues, and improve platform accuracy and performance. When audio is used for training, Modulate extracts acoustic and linguistic patterns to improve model performance. Modulate may also create aggregated or de-identified data from platform usage and submitted audio, kept in a form intended not to identify you or any speaker. This data may be retained and used after your account closes. Customer audio is never sold or used for purposes unrelated to providing and improving the platform. What data does Modulate collect about my API usage? Modulate collects account identifiers, API usage logs, session data, and any metadata you provide. Usage metadata, such as timestamps, API call counts, and conversation counts, is kept separately from audio. It may be retained as long as needed for billing, analytics, security, and legal compliance. API logs are also used to detect abuse, enforce usage limits, improve the platform, and produce aggregated analytics. Who owns the outputs? You retain ownership of the audio you upload. Modulate owns the outputs the platform generates and grants you a license to use them for your internal business purposes. Who is responsible for privacy compliance when I use the API to process my users’ audio? When you use the Modulate platform to analyze audio from your own end users, you are the data controller and Modulate acts as a data processor on your behalf. This means you are responsible for:

Providing required privacy notices to your end users before collecting audio
Obtaining any necessary consents for recording and analysis
Establishing a lawful basis for processing under applicable data protection laws
Complying with audio recording laws in your jurisdiction (wiretapping statutes, consent-to-record requirements, biometric data regulations)
Responding to your end users’ data rights requests (access, deletion, correction)

If your end users submit data rights requests related to audio you processed through the API, those requests should come to you as the data controller. You can then request Modulate’s assistance, and Modulate will cooperate with verified requests. Can I send audio containing health information? If your audio contains protected health information subject to HIPAA, you must have a Business Associate Agreement (BAA) in place with Modulate before uploading it. Contact legal@modulate.ai to set one up. Does Modulate offer a Data Processing Agreement (DPA)? Yes. If you are subject to GDPR, UK GDPR, or other regulations requiring a DPA, Modulate offers a standard agreement that includes appropriate data protection terms, security commitments, and Standard Contractual Clauses for international transfers. Contact legal@modulate.ai to request a DPA. Where is my data processed and stored? Audio and account data are primarily processed and stored in the United States. For customers transferring data from the EU, UK, or other jurisdictions with cross-border transfer requirements, Modulate implements appropriate safeguards including Standard Contractual Clauses. Will Modulate share my data with law enforcement? Modulate may disclose submitted content when required by law, subpoena, court order, or other legal process. It may also disclose content when reasonably necessary to prevent illegal activity, fraud, abuse, or harm. Modulate will notify you of such requests unless legally prohibited. How do I exercise my privacy rights or submit a data request? For requests about your own account data, contact privacy@modulate.ai. Modulate aims to respond within 30 days. If you are a customer responding to one of your end users’ rights requests and need access to data in Modulate’s systems, contact support@modulate.ai with verification of the request. To opt out of marketing communications, click the unsubscribe link in any Modulate email or contact privacy@modulate.ai directly. What happens to my data if I close my account? Stored content is deleted under the retention policies in the Velma Terms. Modulate’s right to use your content continues only as long as needed to finish processing already in progress and to meet retention obligations. Aggregated or de-identified data may be kept after closure. Export anything you want to keep before you close your account.

This section summarizes the terms most relevant to API users. The Velma Terms govern if this FAQ and the Terms conflict. For complete information, including cookie practices, third-party data sharing, and regional rights, see Modulate’s full Privacy Policy and the Velma Services Privacy Addendum.

Support

How do I get help? Check out our Support page. Where can I learn more about Modulate? Visit modulate.ai to learn about Modulate’s mission, or try Velma Preview to explore voice analysis in the browser without writing any code.

​Authentication and API keys

​Models and capabilities

​Audio formats and file requirements

​Pricing and billing

​Rate limits

​Streaming (WebSocket)

​Errors

​Privacy and data

​Support

Authentication and API keys

Models and capabilities

Audio formats and file requirements

Pricing and billing

Rate limits

Streaming (WebSocket)

Errors

Privacy and data

Support