Skip to content

PyAI Hear

American English speech to text

American English speech to text converts spoken American English into a written transcript. Use streaming recognition during a call or recorded transcription when you already have the audio.
· Coming soon — availability unconfirmed; no launch date announced

Prepare your workflow

This is an educational preview. This language and feature combination is not confirmed in the current catalog. Confirm availability before integrating it.

How American English speech to text works

Speech to text converts spoken audio into written words. Streaming recognition produces revisable partial text during a conversation; recorded transcription processes an uploaded recording.

Turn customer calls into searchable transcripts, useful summaries and reviewable quality evidence. Use Hear when your application needs text from speech. Keep streaming partials separate from final transcripts, and use speaker diarization on supported recorded-transcription jobs when you need speaker turns.

How PyAI approaches speech to text

Quality on the actual call

Evaluate the transcript against the original recording, including names, amounts, corrections and language switches. Preserve the original channel layout. Hear streaming accepts mono 8 kHz PCM16; recorded files must have an accurate container and sample rate.

A clear integration path

Use Hear: file transcription for text, asynchronous transcription jobs for speakers and timestamps, or streaming STT for live captions. Keep API keys on your backend.

Predictable usage billing

Review the product meter, sandbox limits and managed telephony terms on the pricing page. Transcription consumes Hear usage. Check identity, available credit, limits and retention before submitting a recording.

Published reference: ~200 ms Hear first partial, in-region. Source: Live in-region, streaming partials; benchmark snapshot 2026-06-24. These are in-region reference measurements, not a language-specific result or latency guarantee. Network, load and deployment affect what you observe. Read the measurement context.

Steps to evaluate when support is available

  1. 01

    Choose recorded transcription or a realtime Hear stream for your audio.

  2. 02

    Select a supported language or automatic detection where available, and supply relevant vocabulary.

  3. 03

    Use final transcript text in your workflow; review critical names, amounts and corrections against the audio.

Put it to work

Call review

Preserve the words, corrections and outcome a reviewer needs to understand what happened.

Compare on the workflow you need

Comparing PyAI with Deepgram? Use the same noisy phone recordings and score business details as well as word error rate. Measure first partial and final text separately.

Read the comparison and migration guidance

American English speech to text FAQ

What is American English speech to text?

American English speech to text converts spoken American English into a written transcript. Use streaming recognition during a call or recorded transcription when you already have the audio.

Does PyAI support American English speech to text?

American English speech to text is not confirmed in the current product catalog. No launch date is announced.

How do I transcribe American English calls?

Use POST /v1/audio/transcriptions for a simple file transcript, POST /v1/transcription/jobs for recorded speaker labels and timestamps, or the Hear WebSocket for live partials and finals. None of these requests generates speech.

Can I get speaker labels and timestamps?

Request diarize: true on a mono async job, or channel: true when the participants are recorded on separate channels. Inspect result.segments and result.words; speaker labels may be absent if diarization falls back. Synchronous verbose_json alone does not guarantee these fields.

How do I handle Hindi-English switching?

For mixed Hindi-English calls, omit the spoken-language hint and evaluate names, amounts, negations and language switches against the recording. Async transcription_language: hi is for known Hindi speech; output_script: latin changes the writing system, not the spoken language or meaning. No Hinglish accuracy guarantee is established.

What does transcription cost and where does audio go?

Hear processes audio in the cloud. Verify project identity, permission, budget and retention before uploading. Async uploaded audio may remain for up to 7 days and results for up to 30 days; there is no zero-retention flag. See https://pyai.com/pricing for current transcription billing.

Can I keep my existing transcription client?

Hear provides an OpenAI-compatible file-transcription API with model pyai-hear. Async jobs and live streaming use their own documented contracts. Select transcription language and output fields rather than voice or synthesis settings.

Current pricing and billing terms

Confirm the next step for your language