Convert and Record Audio in Studio

• Studio is a Pro feature.
• Add or download the AI voice you want before opening the conversion step.
• Stop live voice conversion before recording or converting in Studio.
• Choose the physical microphone you want to record in Settings > Audio.

1. 1

   Play the input

   Use the play button beside the waveform to confirm the source audio is correct.

2. 2

   Move the start handle

   Drag the left trim point past silence or material you do not want converted.

3. 3

   Move the end handle

   Drag the right trim point to set where conversion should stop.

4. 4

   Preview the selection

   Play again and confirm the selected range before starting a longer conversion.

1. 1

   Select an installed AI voice

   Studio loads its own model session and shows when the selected voice is ready.

2. 2

   Set pitch

   Start from the pitch saved for that voice, then adjust only if the source and target vocal ranges need it.

3. 3

   Choose WAV or MP3

   Use WAV when you want an uncompressed editing source. Use MP3 for a smaller, convenient listening or sharing file.

1. 1

   Click Convert

   The button becomes available when input audio is loaded, the voice is ready, and live mode is stopped.

2. 2

   Watch progress

   Keep Echo open while Studio processes the selected range. Longer files and slower backends take more time.

3. 3

   Review the result

   Studio shows the output duration, sample rate, and format when conversion completes.

• Use the output play control to listen inside Studio.
• Seek through the result to check more than the opening seconds.
• Click Save File and choose where to export the finished WAV or MP3.
• Keep the original source until you have confirmed the saved result.

Studio creates an independent conversion session and releases its model and media state after each run. Repeated conversions should not become progressively crackly, wobbly, or distorted.

1. 1

   Confirm whether the same input also changed

   Replay the source. If only later outputs degrade, keep the affected outputs and note the conversion number.

2. 2

   Try a short repeat with the same settings

   Use the same voice, pitch, input range, and output format so the report has a controlled comparison.

3. 3

   Collect support details

   Include Echo version, backend, voice name, input format and length, conversion count before failure, and the exact visible error.