Skip to content

updated the stt streaming code to track and calculate message latency #6

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
jjmaldonis wants to merge 1 commit into
base: main
Choose a base branch
from
Open

updated the stt streaming code to track and calculate message latency #6

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
59 changes: 44 additions & 15 deletions stt_stream_file/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,12 +38,27 @@ uv run stream_audio_file.py --ui --live \

### Save & Print Mode

**Stream to file (realtime mode - natural pace):**
**Stream and save JSON output:**
```bash
uv run stream_audio_file.py -o output.json -f audio.wav --realtime \
uv run stream_audio_file.py -f audio.wav \
--url "wss://api.deepgram.com/v1/listen?model=nova-3&interim_results=true"
```

Output is automatically saved to `audio.json` (derived from input filename).

**Specify a custom output file:**
```bash
uv run stream_audio_file.py -o output.json -f audio.wav \
--url "wss://api.deepgram.com/v1/listen?model=nova-3&interim_results=true"
```

**Live recording saves with timestamp:**
```bash
uv run stream_audio_file.py --live \
--url "wss://api.deepgram.com/v1/listen?model=nova-3&interim_results=true"
# Saves to recording_20250114_153022.json (or similar)
```

**Print basic transcript:**
```bash
uv run print_transcript.py -f output.json
Expand Down Expand Up @@ -74,24 +89,38 @@ It knows this because it knows where it isn't.
## Key Options

### stream_audio_file.py
- `--ui` - Interactive terminal UI with live updates
- `-f, --audio` - Audio file to stream
- `-l, --live` - Stream from microphone
- `-o, --output` - Save JSON messages to file
- `-v, -vv, -vvv` - Increase verbosity

| Option | Description |
|--------|-------------|
| `--url, -u` | Deepgram websocket URL (required) |
| `--ui` | Interactive terminal UI with live updates |
| `-f, --audio` | Audio file to stream |
| `-l, --live` | Stream from microphone |
| `-o, --output` | Save JSON messages to file (defaults to input filename or timestamped name) |
| `-v, -vv, -vvv` | Increase verbosity |

### print_transcript.py
- `--print-speakers` - Show speaker labels
- `--print-channels` - Show audio channels
- `--print-interim` - Include interim results
- `--print-delay` - Show latency (time since end of speech)
- `--colorize` - Color words by confidence
- `--only-transcript` - Just the text, no metadata

| Option | Description |
|--------|-------------|
| `--print-speakers` | Show speaker labels |
| `--print-channels` | Show audio channels |
| `--print-interim` | Include interim results |
| `--print-delay` | Show latency (time since end of speech) |
| `--colorize` | Color words by confidence |
| `--only-transcript` | Just the text, no metadata |

Run either script with `--help` for full options.

## What's Happening?
### Shell Completion

Generate shell completions for your preferred shell:

The UI mode shows transcription speed in real-time - watch words appear as you speak and see exactly how fast Deepgram processes your audio. The `--print-delay` option reveals latency metrics, perfect for testing different models and configurations.
```bash
uv run stream_audio_file.py completion bash # or zsh, fish
```

## What's Happening?

The UI mode shows transcription speed in real-time—watch words appear as you speak and see exactly how fast Deepgram processes your audio. The `--print-delay` option reveals latency metrics, perfect for testing different models and configurations.

Loading