Repository navigation
Features Audio
Every tool built on the onboard microphone and I2S speaker. Driven by two firmware
streams: a continuous raw-PCM mic stream (mic.pcm, 16 kHz mono) and streamed PCM speaker
playback (spkr_pcm_begin / spkr_pcm_data / spkr_pcm_stop). All the DSP runs on the
phone; the device just captures and plays. Gated on caps.mic (input) and caps.speaker
(output).
Live FFT frequency bars computed from the device mic stream.
A scrolling time x frequency spectrogram, drawn from the audio over time.
Decodes received audio into an image (Robot36, Robot72, Martin M1/M2, Scottie S1/S2) and transmits a chosen image out the speaker (Robot36 and more).
Detects the dominant frequency from the live mic, generates tones through the device speaker, and has a full DTMF keypad and decoder:
- DTMF keypad - tap a key and the device synthesises the dual tone straight to its I2S speaker (full-scale, the gain carries the volume, exactly like the RTTTL song path), so it is as loud and audible as the songs. The sound comes out of the ESP, not the phone.
- DTMF decoder - press Start listening to put the ESP mic in listening mode; the PCM is streamed to the phone, which runs a Goertzel decoder and shows the keys it hears live on screen, with a mic-level meter. Press Stop to end. Mic in on the device, decode and display on the phone; a short silence gap separates repeated presses of the same key. Hold an external DTMF source (another phone, a call) near the device mic - the device cannot decode its own keypad, because the mic and speaker are half-duplex (the mic only runs while listening, so the speaker tools are free otherwise).
A live time-domain waveform with an edge trigger, adjustable time base and gain, and Vpp / RMS / frequency readouts.
Plays meme RTTTL melodies natively on the device, now running in the background so Stop actually interrupts them. Speaker volume is controllable from the phone.
Build your own melody note-by-note in the Tone tool: pick a note length and octave, tap notes (and rests), set the tempo, then play the tune on the device speaker. It is assembled on the phone into an RTTTL string and sent to the same background player - no new firmware. Composed songs can be named and saved; they survive app restarts and appear under "My songs" for one-tap replay.
Per-tool recording captures a WAV plus spectrum/waterfall/oscilloscope PNGs for later
review, and the Tools page's microphone recorder (mic_record) saves a WAV straight to
the SD card.
Back to Features.